Soil Geochemistry and Contamination Zoning in Northeastern Ghana: Insights From the Bongo and Talensi Districts

Soil Geochemistry and Contamination Zoning in Northeastern Ghana: Insights From the Bongo and Talensi Districts | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Soil Geochemistry and Contamination Zoning in Northeastern Ghana: Insights From the Bongo and Talensi Districts Belinda S. Berdie, Raymond W. Kazapoe, Darwin A. Awog-Badek, Blestmond A. Brako, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7819579/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Jan, 2026 Read the published version in Environmental Geochemistry and Health → Version 1 posted 7 You are reading this latest preprint version Abstract Understanding natural geochemical background levels is essential for distinguishing anthropogenic pollution from lithogenic signatures in environmental studies. This research focuses on soil geochemical compositions in the Bongo and Talensi districts of northern Ghana, where limited geochemical characterization has hindered effective environmental assessments. This study applies advanced geostatistical methods including Iterative Frequency Distribution and Concentration-Area techniques, by integrating traditional geochemical analyses with multifractal modelling, to provide reliable baseline data for environmental management, land use planning, resource mapping. This study establishes regional geochemical background indicators and baseline values while employing machine learning techniques to evaluate geochemical indicators and delineate spatial geochemical zones. Results reveal distinct geochemical provinces, with Talensi showing increased trace metal concentrations and geochemical clusters linked to volcanic and possibly hydrothermal influences, while Bongo is characterized primarily by silicate weathering processes and minimal trace element variability. Chromium (Cr) emerges as the dominant trace element, with concentrations surpassing international threshold levels, indicating lithogenic enrichment rather than human-induced contamination. While Cu, Pb, Zn, Co, and As levels remain within regulatory standards, variations in Ni and V highlight the region’s heterogeneous bedrock influences. The findings demonstrate that natural geochemical variations – rather than direct anthropogenic pollution – primarily control soil chemistry in these districts. Concluding, the study offers crucial insights into geochemical spatial distribution while contributing to improved strategies for pollution assessment, mineral exploration and sustainable environmental governance in Ghana. Soil Geochemistry Geochemical background levels multifractal modelling machine learning trace element distribution northern Ghana Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 1. Introduction Soil geochemical studies require knowledge of natural chemical element concentrations to differentiate between anthropogenic pollution from lithogenic signatures in environmental frameworks. This foundational concept, termed geochemical background, has its origins in exploration geochemistry, where it was traditionally used to delineate mineral anomalies from natural levels of chemical constituents in geological units (Hawkes & Webb, 1962 ; Abimbola & Olatunji, 2011 ). Environmental geochemistry now incorporates geochemical background as a core principle which enables researchers to study both natural system variations and human-induced impacts on Earth surface chemicals (Filzmoser et al., 2005 ). Geochemical background is defined as the normal concentration range of elements in environmental media that are unaffected by human activity. The term geochemical baseline which gained popularity through International Geological Correlation Programme initiatives depicts a momentary element concentration measurement but may contain anthropogenic component stain from sources (Salminen & Tarvainen, 1997 ). The baseline provides understanding of both natural background levels and weak effects from human activities and airborne pollutants while presenting improved temporal analysis opportunities according to Johnson et al. ( 2005 ) and Reimann et al. ( 2018 ). Despite its recognized significance, studies aimed at systematically determining geochemical background and baseline values remain scarce in Ghana and, more broadly, across West Africa (Amuah et al., 2024 ; Kazapoe and Arhin, 2021 ). This research gap is particularly evident in northern Ghana, where regions such as the Bongo and Talensi districts of the Upper East Region exhibit complex geological settings with limited geochemical characterization. Talensi is particularly heavily influenced by artisanal and small-scale gold mining (ASGM), which poses a significant risk of soil contamination and geochemical distortion. Soil contamination assessment becomes complicated when there is no access to verified baseline and background data that help determine specific contamination thresholds needed for proper cleanup efforts (Armah et al., 2014 ; Bempah & Ewusi, 2016 ). The geological framework of Bongo and Talensi is predominantly composed of Paleoproterozoic Birimian rocks and associated granitoids, which are renowned for their metallogenic significance, especially with regard to gold mineralization. Because of various geological formations with related weathered features these districts present complications that may probably interfere with the distribution patterns of elements (e.g. Bayari et al. 2019 ) thus demanding separation between weathered geologic signatures and human-caused environmental alterations. This research addresses the critical data shortage regarding basic geochemical values by evaluating major and trace elements in district soils to identify natural background thresholds. This study therefore combines traditional methods with the multifractal method to determine regional geochemical background values and baseline values in the Talensi and Bongo districts of the Upper East region of Ghana. Machine learning is further used to comprehensively evaluate the geochemical indicator importance through which it delineates geochemical zones across various study areas. This approach enhances our understanding of the spatial patterns and variability of geochemical components across the study area. It provides a scientifically sound foundation for the exploration, evaluation, and development of natural resources. By using this framework for the first time in northern Ghana, essential findings about both geochemical spatial distribution across the area and pollution risks from mining activities can be established. The researched outcomes will add substantial value to academic publications combined with local environmental governance because they can guide future approaches to assessing pollution and land use planning and natural resource management. 2. Materials and Methods 2.1 Study Area The study was conducted in the Bongo and Talensi districts, located in the Upper East Region of Ghana (Fig. 1 ). This geographical area belongs to the savannah ecological zone where semi-arid climate prevails. The Bongo District occupies a geographic position between 10°45′N and 11°10′N and 0°48′W and 1°05′W while sharing western borders with Kassena-Nankana and southern boundaries with Bolgatanga Municipality and northern and eastern edges bordering Burkina Faso (Ministry of Food and Agriculture [MoFA], 2024). The Talensi District extends between 0°30′00″W to 0°45′00″W west longitude and 10°35′00″N to 10°50′00″N north latitude. The area borders Bolgatanga District to its north whereas Kassena-Nankana District occupies its west and Bawku West District stands to its east while West and East Mamprusi districts position themselves to the south (MoFA, 2024). The geology of both areas shows Palaeoproterozoic lithologies belonging to the Birimian Supergroup (Fig. 2 ). The basement structure of the West African Craton consists of metavolcanic and metasedimentary rocks along with granitoid intrusions according to Griffis et al. (2002) and Kazapoe ( 2023 ). These formations are known to host gold-bearing quartz veins, making the region historically significant for both mechanized and artisanal gold mining activities (Kwayisi et al., 2024 ). The districts show a smooth topographical layout because their slopes range from 1% to 5%. The locations at Tongo and Nangodi and their adjacent steep terrain have slopes reaching up to 10% that trigger occasional soil losses through mass movements and erosion. Granitic and volcanic rock outcrops disrupt the area while influencing soil development and drainage system formation (MoFA, 2024). The area experiences seasonal water drainage from White and Red Volta Rivers through their tributaries. The tributary networks offer limited water resources during rainy months which helps charge underground aquifers that serve as the main water source for both domestic uses and farming operations. Both Bongo and Talensi districts have their economies based primarily on small-scale agricultural farming with additional artisanal mining activities. Among the dispersed rural communities of Bongo District most residents cultivate millet and sorghum. Similarly, the Talensi District, with a population density of about 99 persons per square kilometre, combines dryland agriculture with intensive artisanal mining, particularly in and around Nangodi (Ghana Statistical Service [GSS], 2021; MoFA, 2024). The use of the land under these practices creates pressure which degrades soils and exposes humans to dangerous elements. 2.2 Data Preparation Geochemical datasets were compiled for two study areas Bongo and Talensi, in North-Eastern Ghana (Fig. 1 ). The datasets consist of concentrations for a suite of major oxides and trace elements derived from previously collected soil samples. The datasets were cleaned and appropriate conversions undertaken. 2.3 Determination of Geochemical Background Values Three methods were employed for estimating the geochemical background values of examined elements and compounds. Researchers executed all implementation of these methods through Python 3.10 by employing the “scipy.stats” library to perform statistics along with “matplotlib/seaborn” library to create visualizations. 2.3.1 Method 1: Iterative Method According to Wen et al ( 2024 ) the iterative method served as the calculations basis to determine geochemical background values (GBVs) for study elements in this work. Systematic removal of extreme values helps the iterative method transform non-normal distributions into ones which more closely resemble normal distributions. A computation of mean (X₁) and standard deviation (SD₁) occurs for each element. A set of potential outliers gets identified during the first iteration through a values comparison against X₁ ± 3SD₁ range. The second set of calculations for X₂ and SD₂ uses all remaining values obtained after removing the initial outlier data for the second iteration. The procedure continues toward successive iterations to determine additional outliers until the third computation (X₃ and SD₃) with modified outlier elimination. The process will repeat infinitely until all outliers disappear from the loop (Xₙ ± 3SDₙ). 2.3.2 Method 2: Frequency Distribution The frequency distribution method employed for calculating geochemical background values was also adapted from (Wen et al, 2024 ). Before implementation, the dataset was first subjected to log(x + 1) transformation to reduce right-skewness and normalize distributions with zero or near-zero values. Frequency histograms were then constructed for the transformed data. These histograms helped visualize the modal classes, this is the class with the highest frequency, the spread of values, and any anomalies at the tails. To objectively determine the crowd value (M₀), the method involves projecting two geometric connections across the histogram, firstly, a line from the top-left corner of the modal class to the top-left corner of the next class to the right. Next, another line from the top-right corner of the modal class to the top-right corner of the preceding class to the left is drawn. The point where these two lines intersect and project to the x-axis corresponds to M₀, considered the peak of the distribution. After this, a horizontal line is drawn at 60% of the maximum frequency (0.6f). The intersections of this line with both sides of the frequency curve marks the spread of data around M₀. The distance from M₀ to each of these intersection points defines the mean square error (σ). This σ is then used to establish the lower threshold of geochemical anomalies ( \(\:{M}_{0}+2\sigma\:)\) . Mathematically, M₀ is calculated using: $$\:{M}_{0}=\:{x}_{0}+\frac{i({p}_{2}-{p}_{1})}{2{p}_{2}-{p}_{1}-{p}_{3}}$$ 1 Where \(\:{x}_{0}\) is the lower boundary of the modal class, \(\:i\) the width of class intervals, \(\:{p}_{1}\) the frequency of the class before the modal class, \(\:{p}_{2}\) the frequency of the modal class, and \(\:{p}_{3}\) the frequency of the class after the modal class. 2.3.3 Method 3: (C–A) Method In this study, the C–A method adapted from (Wen et al, 2024 ) was operationalized by constructing log–log cumulative frequency plots of elemental concentrations. The cumulative frequency of the data points exceeding a given concentration threshold was plotted against those concentration values on a log–log scale, yielding a piecewise linear trend that reflects the power-law behaviour of the dataset. For each element, concentrations were averaged from multiple samples then sorted in descending order. The cumulative frequency (rank/total) was computed for each value, and both the concentration and the cumulative frequency were transformed using base-10 logarithm. A double logarithmic plot, this is, log concentration to log cumulative frequency was constructed. In most cases, the data followed a multi-segment linear pattern, interpreted as follows. The first segment which is a steep slope typically reflects anomalously high values, often associated with mineralization or contamination. The second segment which is a gentler slope is assumed to represent the non-singular background population of the element in the environment. A linear regression model is then fitted to the second segment of the curve, and the arithmetic mean of the concentrations within this segment is calculated. This value is then recorded as the geochemical background value C₀₃ for that element. Unlike the frequency or iterative methods, this method does not rely on standard deviation thresholds to remove high values. Instead, it identifies thresholds through fractal breaks in slope on the log–log plot, thereby improving background estimation in non-Gaussian datasets. Having determined geochemical background values using all three methods, the final geochemical background value, C₀ for each element was then obtained by averaging the values calculated from the three methods this is the Averaged Geochemical Background Values (AGBV). 2.4 Determination of Geochemical Baseline Values (GBV) and Normalization The baseline values, however, were determined using a distribution driven classification method and applied to the original dataset. This process begins with testing the data for normality, to classify the distribution of each element across the dataset. If the dataset conforms to a normal distribution, the arithmetic mean plus or minus two standard deviations is used to define the baseline value and its range. If the dataset adheres to a log-normal distribution, then the geometric mean is multiplied and divided by the square of the geometric standard deviation (GSD²) to determine the baseline range. A geometric formulation maintains statistical consistency between skewed distributions that do not follow normal or log-normal patterns by taking asymmetry into account. The GBVs received their standardization treatment through UCC values adopted from Wedepohl ( 1995 ) and Taylor ( 1964 ). The GBV/UCC and Background-to-Baseline Ratio measurements helped determine elemental excess levels compared to natural rock composition for detecting human sources of contamination. Factors exceeding one value suggested that the elements exceeded their natural distribution in the environment. 2.5 Cluster Analysis The hierarchical cluster analysis (HCA) assessed elemental distribution patterns by utilizing background values of trace elements collected from Bongo and Talensi. The unguided learning approach examined element distribution patterns to determine behavioural clusters between geochemical entities and locate distinctive compositional signatures of each location. The analysis selected trace elements exclusively to prevent major oxide scale bias that occurs because these elements exist at much higher concentrations. Z-score normalization standardization techniques applied to data allowed element comparisons despite various unit differences and magnitude levels. The dataset transformation produced a unitless scale that positioned each variable at zero standard deviation while maintaining a measurement scale of one which created equal importance for all trace elements during clustering. Euclidean distance measures the dissimilarities between clusters while Ward’s linkage method executes the clustering process. A Heatmap Visualization of Dendrogram displayed the HCA output. 2.6 Machine Learning for Feature Importance A supervised machine learning technique served to evaluate additional aspects about geochemical indicator differences between the Bongo and Talensi regions. The method enabled the determination of the relative prominence of each major and trace element in establishing regional classifications while strengthening geochemical zoning strategies. The input dataset consisted of the original dataset for both regions, with elements and compounds serving as independent variables (features) and the region (Bongo or Talensi) as the dependent variable (target). The regions were one-hot encoded to facilitate regression-based modelling. All features were standardized using Z-score normalization* to ensure uniform influence across variables of differing scales and units. 2.7 Model implementation A regression model was employed as the primary predictive algorithm. To increase its robustness, Lasso (L1) and Ridge (L2) regularization were implemented through Elastic Net models were tested for comparison. This helps minimize overfitting, a common risk when working with high-dimensional geochemical data and a limited number of classes. Model performance was evaluated using R² and Adjusted R² for explained variance, Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) for prediction accuracy. Cross-validation (k-fold) was also implemented to ensure model generalizability. Lastly, a hyperparameter tuner was employed using GridSearchCV to identify the optimal set of parameters for the model variant. The absolute values of the regression coefficients from the best-performing model were extracted to estimate the relative feature importance of each element. Higher absolute coefficient values indicated a greater role of that element in distinguishing between the two regions. 3. Results 3.1 Geochemical Background Values Most of the elemental indicators achieved internal consistency from background values estimated through Iterative 3σ, Frequency Histogram and Concentration-Area (C–A) methods. Geochemical distribution patterns throughout the Bongo and Talensi regions demonstrate a stable central value presence yet show medium-level effects from distributional skewness in addition to outlier-induced variations, particularly for Cd, Sb and Ag trace elements. The minor differences in the method stem from differences in tail effects sensitivity and distribution assumption sensitivity yet match Reimann and Garrett’s ( 2005 ) view that no single method operates on all geochemical data types since natural systems lack normality. 3.2 Baseline Values The baseline values were calculated by using the arithmetic mean ± 2SD for normal distribution elements and the geometric mean for elements following a log-normal distribution. The baseline ranges for elements Cd and Ge required adjustment because these elements show substantial skewness and have naturally high concentration patterns across lithologic units. The separation of natural and anthropogenic activities is essential because hazardous mining effects may amplify the existing geological variations in Talensi. The baseline calculations (Fig. 3 ) indicated that elemental levels in most tested samples maintained typical natural concentrations, demonstrating that the studied areas' geochemical status mostly remained pristine until elements showed increased patterns in specific loci. 3.3 UCC and Baseline Ratios Figure 4 shows the background to UCC enrichment factors for the Bongo district. Comparison with Upper Continental Crust (UCC) values (Wedepohl, 1995 ) highlighted pronounced regional differences in elemental enrichment. In the Talensi domain (Fig. 5 ), the ratios of Cd, Sb and Ag surpassed unity, whereas other elements remained below unity, which could result from natural or human disturbances to the area. The measured levels of elements likely result from aspects of sulfide mineralisation properties, together with natural baseline variations and former mining activities across West African Birimian regions (BGS, 1996). The geochemical analysis shows SiO₂, K₂O, and Al₂O₃ stay under UCC values or align within UCC constraints in Bongo, particularly because of its felsic and weathered geologic origin with negligible metallogenic enhancements. Geological studies confirm that the Bongo area contains granitic gneiss rocks, which contrast with Talensi (Fig. 6 ) where metavolcanic belts hold the potential for Au-As mineralisation. 3.4 Geochemical Grouping using Ternary Diagrams The three ternary diagrams of (SiO₂/10–Al₂O₃–CaO), (K₂O + Na₂O–CaO–TFe₂O₃) and (CaO–K₂O–Na₂O) from Fig. 7 – 9 respectively, show separate groupings which demonstrate the distinct mineralogical and geochemical settings of Bongo and Talensi regions. Bongo rocks accumulate in the silica-rich end-member on the SiO₂/10–Al₂O₃–CaO diagram (Fig. 7 ) because they contain quartz and show weathering features. Talensi samples show higher proportions of CaO and Fe₂O₃ components when comparing equivalent elemental proportions because their mineral compositions are heavily influenced by ferromagnesian silicates as well as carbonates. Talensi samples in the (K₂O + Na₂O)–CaO–TFe₂O₃ diagram (Fig. 8 ) exhibit higher TFe₂O₃ proportions that match mafic source rock weathering and mineralised formations. Soil chemistry across the region exhibits these geochemical associations because they result from the interaction between lithological variations, mineral weathering mechanisms and metal transport patterns in the area (Reimann et al., 2008 ). 3.5 Cluster Analysis The dendrogram structure generated from cluster analysis divides the elemental data into two major domains corresponding to the geographical extent of Bongo and Talensi (Fig. 11 ). The group of trace elements found in the Talensi cluster shows a uniform geochemical signature that comes from common mineral sources or substance transport routes which link to ore deposits and human-made soil disturbance. The Bongo cluster consists of lithophile elements that initially came from felsic and less metalliferous rock. Further exploration is justified at sub-regional levels because geochemical region-specific zoning patterns have been identified while also benefiting environmental monitoring and mineral exploration needs. 3.6 Machine Learning for Feature Importance 3.6.1 Model Evaluation The model performance metrics indicated a moderately strong fit, suggesting that geochemical background values hold substantial predictive power for regional classification. The results suggest that approximately 66% of the variance in regional differentiation could be explained by the model, with prediction errors kept within a relatively narrow margin. The regression model identified several geochemical variables as key contributors to regional differentiation between Bongo and Talensi. Feature importance (Fig. 11 ) was assessed based on the absolute value of the standardised regression coefficients, with higher values indicating greater influence on the classification outcome. Performance metrics acquired from using the regression model demonstrate a moderate level of fit with R² = 0.6623 and Adjusted R² = 0.5675 in addition to RMSE = 0.2778 and MAE = 0.2029. The elemental predictors account for sixty-two percent of the variation that exists in regional geochemical differentiation patterns. Using standardised regression coefficients, we determined the feature importance where Cd emerged as the primary variable and Sb and Ge followed behind, while Ag ranked fourth. Multiple trace metals and metalloids rank foremost among top contributors, which signifies that region-type determination relies mostly on elements which naturally associate with mineralisation and show heightened sensitivity to earth and anthropogenic sources. Research in identical African locations has proven that heavy metals operate as effective markers to track both geological wealth potential and human-caused environmental damage (Obiri et al., 2016 ). The regression model proved effective, yet it requires recognition of its insufficient capability to handle specific situations. Other non-linear influencing factors that influence elemental concentrations include redox conditions and micro-topography, and mineral grain size because they are not fully represented in the prediction model. 3.7 Discussions Collectively, the geochemical, statistical, and machine learning results converge on a robust delineation of geochemical provinces within the Upper East Region of Ghana. The Talensi area displays increased trace metals as well as distinct geochemical clusters combined with volcanic and possibly hydrothermal indicators in its mineralogy. The Bongo region displays geochemical indicators consisting mainly of silicate weathering materials in addition to minimal trace elements and limited alteration of chemical signatures. The analysis demonstrated that Cr appeared as the dominant trace element, reaching 276.57 mg/kg in the final average. Zr and strontium Sr followed, with concentrations of 1168.17 mg/kg and 371.88 mg/kg, respectively. The V levels in the samples were 29.07 mg/kg but Cu and Pb levels were lower at 7.87 mg/kg and 4.82 mg/kg, respectively. The elevated Cr concentrations suggest a significant contribution from mafic to ultramafic bedrock sources, corroborating interpretations of a lithological influence on soil chemistry (Taylor & McLennan, 1995 ). The coherence among the three baseline methods was generally high, although V showed substantial variation, as its results varied from 13.01 to 39.64 mg/kg. Such variation is expected given that the (C–A) plot method, in particular, tends to highlight localised enrichments associated with mineralised zones as described in Cheng ( 2007 ). The studied concentrations of Cu and Pb, and Zn remained under the established international standards for critical thresholds. The study showed Bongo soils contained less Cu than two worldwide benchmarks – the Canadian Council of Ministers of the Environment guideline, which amounts to 63 mg/kg, while the United States Environmental Protection Agency residential limit stands at 1600 mg/kg (CCME, 2007; USEPA, 2018). Similarly, Zn concentrations remained far below the 200 mg/kg CCME guideline. The Cr concentration in the Bongo soils exceeded both the CCME threshold of 64 mg/kg and the VROM threshold of 100 mg/kg, which reveals a geogenic natural enrichment process instead of human-induced pollution. This enrichment is consistent with globally reported behaviour of Cr in tropical lateritic profiles developed over ultramafic bedrock (Kabata-Pendias, 2011 ). The analysis revealed that Talensi showed higher Mn levels at 77.04 mg/kg in addition to Zn levels at 26.15 mg/kg as compared to Bongo (Fig. 12 ). The measured Cr content was 185.12 mg/kg in Talensi soils and exceeded the CCME and VROM soil quality standards yet remained lower than Bongo values. Notably, Co averaged 19.56 mg/kg, while Ni concentrations showed high variability, with values ranging between 0.25 and 17.59 mg/kg depending on the method used. The high variability in Ni measurements indicates that either particular concentrations exist in specific locations or that chemical substance variability in the native parent materials happens across the area. The concentrations of Cu and Pb were reported as 22.06 mg/kg and 4.76 mg/kg, and As concentrations averaged 0.82 mg/kg throughout the Talensi area, where all values persisted under the USEPA residential soil guidelines. When benchmarked against international standards, Bongo District soils exhibited mixed trends. The high Cr levels in the soil match both CCME and VROM soil quality limits, thus proving the lithogenic enrichment theory. Soils contained substantial amounts of vanadium, although their levels fell below the World average values at 42.5 mg/kg while remaining at normal background ranges. The Ba content in the Bongo District soils reached 1038.37 mg/kg beyond the Tanzanian Bureau of Standards (TBS) guideline of 100 mg/kg but is similar to standard values for felsic rock-derived soils (Kabata-Pendias, 2011 ). The soil's analytical results showed that Pb, Cu, Zn, Co and As levels remained under both Tanzanian Bureau of Standards limits and international intervention thresholds, while Pb concentrations exceeded TBS guidelines. The Cr content in Talensi soil matched the distribution patterns found in Bongo within international standards. Additionally, the Cr measurements stood above both CCME (64 mg/kg) and VROM (100 mg/kg) thresholds. Soil measurements of V reached or slightly exceeded the World natural background standards. The analysed Ba content reached 465.25 mg/kg, which exceeded the TBS threshold yet maintained values found in mineral-rich soils according to Kabata-Pendias ( 2011 ). The environmental health risks were minimal because Zn, Pb, Cu, Co and As content did not surpass allowable thresholds. The geochemical characteristics of Bongo show predominantly geogenic influences on element distribution patterns, mainly for Cr and Ba. The detected elements' levels indicated no major human-made contamination because they remained low at various sampling sites. Intense lateritization zones exist in the region based on moderate to elevated concentrations of Al₂O₃ and Fe₂O₃, which enhance the potential for finding mineralised structures below the surface. The geographically distributed Cr and Ba patterns reveal separate rock formations, which will help improve upcoming environmental examinations and mineral resource evaluations for the district (Dill, 2010 ). Varying levels of trace elements exist more broadly throughout Talensi territorial space as compared to Bongo. Higher concentration variability of V, Ni and Zn demonstrates that the soil chemistry is influenced by more heterogeneous bedrock in addition to surficial processes. The elevated Cr and V levels within the area most likely originate from specific units of the greenstone belt in addition to weathered ultramafic rock formations. The heterogeneous geochemical conditions require site-specific baseline assessments instead of generalized regional approaches (Taylor & McLennan, 1995 ) because they impact environmental risk evaluations and mineral exploration activities. Conclusion The geochemical baselines established for the Bongo and Talensi districts of northeastern Ghana illustrate a predominantly lithogenic control over soil chemistry, particularly concerning Cr and Ba concentrations. While Cu, Pb, Zn, Co, and As levels remain low, natural enrichments in chromium and barium necessitate careful consideration in land use planning and environmental assessments. Advanced combination methods between Iterative and Frequency Distribution and Concentration–Area have produced reliable baseline data that serves as guidance for upcoming environmental management and mapping operations. This research establishes that natural geochemical variations control the soil composition of these districts more than human-made contaminations do. Declarations Author Contribution Belinda S. Berdie conceptualized the study, led the fieldwork and data analysis and co-authored the manuscript. Raymond W. Kazapoe handled geostatistical modelling and regional geological interpretation and co-authored the manuscript. Darwin A. Awog-Badek handled the machine learning implementation and visualisation. Blestmond A. Brako, Gordon Foli and Simon Y. Gawu supported literature review, provided supervisory guidance and critical manuscript refinement. All authors reviewed the final manuscript and approved its submission. Funding This research was supported by the Ghana Education Trust Fund (GETFUND), and the Ghana National Petroleum Commission (GNPC) Foundation. Institutional support, including laboratory facilities and supervision was provided by C. K. Tedam University of Technology and Applied Sciences (CKT-UTAS), Kwame Nkrumah University of Science and Technology (KNUST), University of Ghana (UG) and the Ghana Geological Survey Authority (GGSA). The authors gratefully acknowledge these organisations for their role in facilitating the successful completion of this study. Conflict of Interest The authors declare no conflict of interest. Data Availability Statement Data will be made available upon reasonable request from the corresponding author. References Abimbola, A. F., & Olatunji, A. S. (2011). Urban geochemical mapping in Nigeria with some examples from southern Nigeria. Mapping the Chemical Environment of Urban Areas , 570-580. Amuah, E.E.Y., Fei-Baffoe, B., Kazapoe, R.W., Dankwa, P., Okyere, I.K., Sackey, L.N.A., Nang, D.B. and Kpiebaya, P. (2024). From the ground up: Unveiling Ghana's soil quality crisis and its ecological and health implications. Innovation and Green Development , 3(1), p.100097. Armah, F. A., Quansah, R., & Luginaah, I. (2014). A systematic review of heavy metals of anthropogenic origin in environmental media and biota in the context of gold mining in Ghana. International scholarly research notices , 2014(1), 252148. https://doi.org/10.1155/2014/252148 Bayari, E. E., Foli, G. and Gawu, S. K. Y. (2019). Geochemical and pathfinder elements assessment in some mineralised Regolith profiles in Bole-Nangodi gold belt in north-eastern Ghana. Environmental Earth Sciences , 78(9), 1-16. https://doi.org/10.1007/s12665-019-8265-5 Bempah, C. K., & Ewusi, A. (2016). Heavy metals contamination and human health risk assessment around Obuasi gold mine in Ghana. Environmental monitoring and assessment , 188, 1-13. https://doi.org/10.1007/s10661-016-5241-3 British Geological Survey (BGS). (1996). Geochemical Atlas of Ghana . Keyworth, Nottingham, Uk.. Canadian Council of Ministers of the Environment (CCME). (2007). Canadian soil quality guidelines for the protection of environmental and human health. Retrieved from https://ccme.ca Cheng, Q. (2007). Mapping singularities with stream sediment geochemical data for mineral exploration. Ore Geology Reviews , 32(1–2), 314–324. Dill, H. G. (2010). The geology of aluminum phosphates and sulfates of the alunite group minerals: An overview. Earth-Science Reviews , 101(1–2), 1–35. Filzmoser, P., Garrett, R. G., & Reimann, C. (2005). Multivariate outlier detection in exploration geochemistry . Computers & geosciences, 31(5), 579-587. Ghana Statistical Service (GSS). (2021). 2021 Population and Housing Census: population of Regions and Districts (Vol. 3A). Accra, Ghana. https://statsghana.gov.gh/gssmain/fileUpload/pressrelease/2021%20PHC% 20General%20Report%20Vol% 203A_Population%20of%20Re gions%20and%20Districts_181121.pdf Hawkes, H.E. and Webb, J.S., 1962. Geochemistry in Mineral Exploration , first edition, Harper & Row, New York. Johnson, C. C., Breward, N., Ander, E. L., & Ault, L. (2005). G-BASE: baseline geochemical mapping of Great Britain and Northern Ireland. Geochemistry: exploration, environment, analysis , 5(4), 347-357. Kabata-Pendias, A. (2011). Trace elements in soils and plants (4th ed.). CRC Press. Kazapoe, R. W. (2023). A review of the characteristics and geological settings of orogenic gold deposits of the Boule Mossi Domain: Implication for gold exploration. Geology, Ecology, and Landscapes , 9(2), 690-705. https://doi.org/10.1080/24749508.2023.2256553 Kazapoe, R. W. & Arhin, E. (2021). Determination of local background and baseline values of elements within the soils of the Birimian Terrain of the Wassa Area of Southwest Ghana. Geology, Ecology, and Landscapes , 5(3), 199-208. Kazapoe, R. W., & Arhin, E. (2019). Selenium in locally produced food crops and implications on healthy eating: A case study at the Talensi District of Ghana . EC Nutrition , 8(3), 85–92. Kesse, G. O. (1985). The mineral and rock resources of Ghana . A.A. Balkema. Kwayisi, D., Kazapoe, R. W., Alidu, S., Sagoe, S. D., Umaru, A. O., Amuah, E. E. Y. & Kpiebaya, P. (2024). Exploring soil pollution patterns in Ghana’s northeastern mining zone using machine learning models. Journal of Hazardous Materials Advances , 16, 100480. Matschullat, J., Ottenstein, R., & Reimann, C. (2000). Geochemical background—can we calculate it? Environmental Geology , 39(9), 990–1000. McLaughlin, M. J., Parker, D. R., & Clarke, J. M. (2000). Metals and micronutrients – food safety issues. Field Crops Research , 60(1–2), 143–163. Ministry of Food and Agriculture. (2024). Upper East Region Agricultural Sector Projects and Potentials . Retrieved from https://mofa.gov.gh/site/directorates/50-regional-directorates/upper-east-region Obiri, S., Dodoo, D. K., & Okai-Sam, F. (2016). Cancer and non-cancer health risk from exposure to arsenic, cadmium, and copper in borehole, tap, and surface water in the Obuasi municipality of Ghana. Human and Ecological Risk Assessment , 22(2), 310–323. https://doi.org/10.1080/10807031003788907 Reimann, C., & Garrett, R. G. (2005). Geochemical background—concept and reality. Science of the Total Environment , 350(1–3), 12–27. Reimann, C., Fabian, K., Flem, B., Andersson, M., Filzmoser, P., & Englmaier, P. (2018). Geosphere-biosphere circulation of chemical elements in soil and plant systems from a 100 km transect from southern central Norway. Science of the Total Environment , 639, 129-145. Reimann, C., Filzmoser, P., & Garrett, R. G. (2008). Statistical data analysis explained: Applied environmental statistics with R . John Wiley & Sons. Salminen, R., & Tarvainen, T. (1997). The problem of defining geochemical baselines. A case study of selected elements and geological materials in Finland. Journal of geochemical exploration , 60(1), 91-98. Taylor, S. R., & McLennan, S. M. (1995). The geochemical evolution of the continental crust. Reviews of Geophysics , 33(2), 241–265. Taylor, S. R. (1964). Abundance of chemical elements in the continental crust: Anew table. Geochimica et Cosmochimica Acta , 28(8), 1273-1285. https://doi.org/10.1016/0016-7037(64)90129-2 United States Environmental Protection Agency (USEPA). (2018). Regional Screening Level (RSLs) Generic Table . Retrieved from https://www.epa.gov/risk/regional-screening-levels-rsls-generic-tables Wedepohl, K. H. (1995). The composition of the continental crust. Geochimica et Cosmochimica Acta , 59(7), 1217–1232. Wen, W., Yang, F., Xie, S., Wang, C., Song, Y., Zhang, Y., & Zhou, W. (2024). Determination of Geochemical Background and Baseline and Research on Geochemical Zoning in the Desert and Sandy Areas of China. Applied Sciences , 14(22), 10612. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 08 Jan, 2026 Read the published version in Environmental Geochemistry and Health → Version 1 posted Editorial decision: Revision requested 03 Nov, 2025 Reviews received at journal 02 Nov, 2025 Reviewers agreed at journal 13 Oct, 2025 Reviewers invited by journal 13 Oct, 2025 Editor assigned by journal 13 Oct, 2025 Submission checks completed at journal 10 Oct, 2025 First submitted to journal 09 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7819579","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":534076869,"identity":"40a6127e-afb4-471f-9b0d-8513b068bee5","order_by":0,"name":"Belinda S. Berdie","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/ElEQVRIiWNgGAWjYBACAwYeIMkmwcDPwIYkyMBwgLAWyQY2xgaGBOK1ABkHiNVizn/24OOKMgs54+PH0h/z/rCzZ2Bv3ibB8OsOTi2WM/KSDc+ckzA2O5N2sJknITmxgedYmQRj3zPcDrvBYybZ2CaRuO0GeyNQC3MCg0SOmQRjz2HcWs6fMf8J1FK/eQZYS709g/wbAloO5JgxArUkGEiwgRx2mLFBgsdMguEHbi0gv0g2nJMwnHEmLXHmnLTjiW08acUWiQ24tYBC7GNDWZ08f/sxgw9vbKrt+dkPb7zx4Q9uLZgAnAgS20jQAQV/SNcyCkbBKBgFwxYAAPqPVLgdJevJAAAAAElFTkSuQmCC","orcid":"","institution":"Kwame Nkrumah University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Belinda","middleName":"S.","lastName":"Berdie","suffix":""},{"id":534076870,"identity":"e7c456f1-0698-4b4a-b70c-c65e455e828e","order_by":1,"name":"Raymond W. Kazapoe","email":"","orcid":"","institution":"University for Development Studies","correspondingAuthor":false,"prefix":"","firstName":"Raymond","middleName":"W.","lastName":"Kazapoe","suffix":""},{"id":534076871,"identity":"0b820a3f-3820-4812-bb61-b9331a38b98a","order_by":2,"name":"Darwin A. Awog-Badek","email":"","orcid":"","institution":"Tamale Teaching Hospital","correspondingAuthor":false,"prefix":"","firstName":"Darwin","middleName":"A.","lastName":"Awog-Badek","suffix":""},{"id":534076872,"identity":"cc5156d8-48b9-4703-92ba-e09b4f63efaa","order_by":3,"name":"Blestmond A. Brako","email":"","orcid":"","institution":"Kwame Nkrumah University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Blestmond","middleName":"A.","lastName":"Brako","suffix":""},{"id":534076873,"identity":"87a72f50-f6e3-451d-96dc-6bba19471dec","order_by":4,"name":"Gordon Foli","email":"","orcid":"","institution":"Kwame Nkrumah University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Gordon","middleName":"","lastName":"Foli","suffix":""},{"id":534076874,"identity":"96ca39dc-86d1-48ec-967b-027644142cad","order_by":5,"name":"Simon K. Y. Gawu","email":"","orcid":"","institution":"Kwame Nkrumah University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Simon","middleName":"K. Y.","lastName":"Gawu","suffix":""}],"badges":[],"createdAt":"2025-10-09 16:08:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7819579/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7819579/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10653-025-02967-y","type":"published","date":"2026-01-08T15:59:10+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":94469944,"identity":"4bcdf943-01af-431c-a3e4-d4bf9664d2d3","added_by":"auto","created_at":"2025-10-27 15:31:09","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1609280,"visible":true,"origin":"","legend":"","description":"","filename":"BSBERDIE1ManuscriptDraft.docx","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/9772ea9355285456d4ee621f.docx"},{"id":94469621,"identity":"bcd6ac11-7975-4e5e-bcea-dedcbca85228","added_by":"auto","created_at":"2025-10-27 15:30:15","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8428,"visible":true,"origin":"","legend":"","description":"","filename":"e76d314e661e458dae87541e20521c5d.json","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/c745285c65b8efaad4d29735.json"},{"id":94469992,"identity":"ad7ebe73-7381-420c-8b45-c04b511a1d17","added_by":"auto","created_at":"2025-10-27 15:31:20","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":90969,"visible":true,"origin":"","legend":"","description":"","filename":"e76d314e661e458dae87541e20521c5d1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/7a3883be219c33f10a71bd9d.xml"},{"id":94470043,"identity":"207c2798-5c74-4866-9b50-abd870c15670","added_by":"auto","created_at":"2025-10-27 15:31:24","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":276115,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/9b3eebaeabea1ca83cc0a8be.jpeg"},{"id":94469609,"identity":"7a06d653-592c-4fc4-9f07-0b27d2737350","added_by":"auto","created_at":"2025-10-27 15:30:07","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":44718,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/1c9710ca082972e3c75d8e7c.png"},{"id":94469912,"identity":"55a85529-8fda-467f-b7f0-08041ee9f2a5","added_by":"auto","created_at":"2025-10-27 15:31:01","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":45395,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/4be9ca799f429a2a404bd395.png"},{"id":94469737,"identity":"5dac2eb5-6747-4e3a-bb79-1a0445b62b19","added_by":"auto","created_at":"2025-10-27 15:30:26","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":234056,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/21f424bcfdfbbdf507516c7d.png"},{"id":94469622,"identity":"a1ecdc27-c942-402f-b125-22a366e03659","added_by":"auto","created_at":"2025-10-27 15:30:18","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":329144,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/3fbad242d22908bf0d79adab.jpeg"},{"id":94469557,"identity":"96dd3261-a061-4544-bae3-8801dffb0d39","added_by":"auto","created_at":"2025-10-27 15:30:01","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":165171,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/903e61153d4d055265c55ffb.png"},{"id":94469796,"identity":"fbb545e3-27af-4ea8-8bb3-bbac6091675d","added_by":"auto","created_at":"2025-10-27 15:30:44","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":125231,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/91bdf90d1b4e5245459bfe31.png"},{"id":94469739,"identity":"033078a8-a352-4784-9435-3d493c6e95ad","added_by":"auto","created_at":"2025-10-27 15:30:26","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":139258,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/62e3f1ce34aec0bb89851b0f.png"},{"id":94469745,"identity":"db53c214-c676-4492-a81d-200f45c1be6e","added_by":"auto","created_at":"2025-10-27 15:30:29","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":131335,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/5a430d3e77ceec9e0c7feb45.png"},{"id":94469603,"identity":"82cb2374-a280-4cdb-ba4f-6b7fd7920418","added_by":"auto","created_at":"2025-10-27 15:30:05","extension":"jpeg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21559,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/5de644e92c8a6223fcc91a93.jpeg"},{"id":94469950,"identity":"d881bd0b-bf12-4014-b6ec-27c26f510c10","added_by":"auto","created_at":"2025-10-27 15:31:11","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22714,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/26ad8bc820b2690e0bd05c48.jpeg"},{"id":94469946,"identity":"e9eba90f-1615-493d-b94b-8df7b3c55dbb","added_by":"auto","created_at":"2025-10-27 15:31:10","extension":"jpeg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21618,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/180ea5dedb44f7a4dd9bb89e.jpeg"},{"id":94470061,"identity":"c90a8706-58c9-40f0-b635-23e71ed0b6d2","added_by":"auto","created_at":"2025-10-27 15:31:25","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":202400,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/3dd7ad224e9419d335860cc9.png"},{"id":94469968,"identity":"1e12ec8c-ae59-48ae-b376-9d4e4d84eb61","added_by":"auto","created_at":"2025-10-27 15:31:14","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21045,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/65f391b89c0f8c72082b020e.png"},{"id":94469917,"identity":"2f146eb4-67c1-4926-a354-a9c46c8a0725","added_by":"auto","created_at":"2025-10-27 15:31:02","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11948,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/c5c955468208243722e6f8f7.png"},{"id":94469610,"identity":"67c093ef-402a-4154-abdf-9864fed5f3dd","added_by":"auto","created_at":"2025-10-27 15:30:08","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":55640,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/69121384be65be411aa5b50b.png"},{"id":94469952,"identity":"b9f838c3-5d5b-4eed-91df-5d5667e7cc8e","added_by":"auto","created_at":"2025-10-27 15:31:11","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":297770,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/bf26d0bd81c9ac542f400310.png"},{"id":94470050,"identity":"1c1929e9-fa0e-4123-ab64-4fc66793c48d","added_by":"auto","created_at":"2025-10-27 15:31:24","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33465,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/55bf7164e33e26aea1fceb67.png"},{"id":94469940,"identity":"9d14557c-9ce3-4417-b115-57e35416f94b","added_by":"auto","created_at":"2025-10-27 15:31:09","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28557,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/d4207aae184a290a1103f763.png"},{"id":94470345,"identity":"4906aaf2-94fd-4c6f-a5d7-3be148182dcc","added_by":"auto","created_at":"2025-10-27 15:31:52","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29684,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/7ee429fe06995722a389b3b8.png"},{"id":94469947,"identity":"35461254-50db-4bc7-b8e8-21b0b134665e","added_by":"auto","created_at":"2025-10-27 15:31:11","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28115,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/132825f945107c08c6c8dc73.png"},{"id":94470054,"identity":"4af30d6b-b9a0-477e-9cfc-f7fc244658e9","added_by":"auto","created_at":"2025-10-27 15:31:24","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6668,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/96df2adb7eda0a32e96fe46c.png"},{"id":94469799,"identity":"e7940a30-a676-4787-a33b-6aabee8c9b2c","added_by":"auto","created_at":"2025-10-27 15:30:44","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6758,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/a6d7c25feb14e71f732f20e4.png"},{"id":94469927,"identity":"3297370c-d570-4569-b6df-83e2f685ba9d","added_by":"auto","created_at":"2025-10-27 15:31:05","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6243,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/5e2ba78e652b820a2e16d6a1.png"},{"id":94470223,"identity":"e3e1a1e1-6d56-4009-8629-4c73352480fc","added_by":"auto","created_at":"2025-10-27 15:31:44","extension":"xml","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":88768,"visible":true,"origin":"","legend":"","description":"","filename":"e76d314e661e458dae87541e20521c5d1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/6da2c69f98e2e6c7838fb95b.xml"},{"id":94469936,"identity":"080d2c5c-bd69-4929-8433-844dc7bfc974","added_by":"auto","created_at":"2025-10-27 15:31:07","extension":"html","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":97969,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/c1117a47b5c432df038bec54.html"},{"id":94469890,"identity":"75406d52-4478-48da-89a9-cd054e893419","added_by":"auto","created_at":"2025-10-27 15:30:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":569654,"visible":true,"origin":"","legend":"\u003cp\u003eA map showing the Bongo and Talensi Districts in the Upper East region of Ghana\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/c2b32a0a084bb9299e3dfba6.png"},{"id":94469560,"identity":"d4ce486d-59c5-403c-8ba5-5908cdff3d19","added_by":"auto","created_at":"2025-10-27 15:30:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":649867,"visible":true,"origin":"","legend":"\u003cp\u003eGeology map of the Bongo and Talensi Districts.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/866212cbe7afa9ba5ffc86d7.png"},{"id":94469978,"identity":"0bb992b5-8041-46db-b87f-13744c208a14","added_by":"auto","created_at":"2025-10-27 15:31:16","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":519018,"visible":true,"origin":"","legend":"\u003cp\u003eChart showing the background to baselines ratios for the Bongo district\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/4e412e8b1765fc12b0ca4a78.png"},{"id":94470207,"identity":"6a9d0c9e-ef47-4190-9843-1e85b204652b","added_by":"auto","created_at":"2025-10-27 15:31:43","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":469646,"visible":true,"origin":"","legend":"\u003cp\u003eChart showing the background to UCC enrichment factors for the Bongo district\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/3f5edfba72fcba3ef16d5052.png"},{"id":94469467,"identity":"a32e22f3-6097-48aa-b9e2-442f73afe0fa","added_by":"auto","created_at":"2025-10-27 15:29:34","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":456523,"visible":true,"origin":"","legend":"\u003cp\u003eChart showing the background to baselines ratios for the Talensi district\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/ae57c64e2a9636455b3f0e2a.png"},{"id":94469913,"identity":"87cafd73-47cb-4f94-a8a2-91f3b6659ab2","added_by":"auto","created_at":"2025-10-27 15:31:02","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":420355,"visible":true,"origin":"","legend":"\u003cp\u003eChart showing the background to UCC enrichment factors for the Talensi district\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/a7366f8816ee6968089fc346.png"},{"id":94469939,"identity":"e2a3077e-c641-4c71-8fe6-0d27cf418610","added_by":"auto","created_at":"2025-10-27 15:31:08","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":38820,"visible":true,"origin":"","legend":"\u003cp\u003eThe SiO₂/10–Al₂O₃–CaO ternary diagram\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/6bf250c937b3cec39982b7dd.png"},{"id":94469933,"identity":"7bc8ee00-2796-4d05-82a2-043c243784ed","added_by":"auto","created_at":"2025-10-27 15:31:06","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":41088,"visible":true,"origin":"","legend":"\u003cp\u003eThe K₂O + Na₂O–CaO–TFe₂O₃ ternary diagram\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/65fe6ae6c9e7fb7a5e330aa6.png"},{"id":94489256,"identity":"220977c3-06c0-4572-bb81-1794ed98d6ae","added_by":"auto","created_at":"2025-10-27 17:04:01","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":36931,"visible":true,"origin":"","legend":"\u003cp\u003eThe CaO–K₂O–Na₂O ternary diagram\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/de9cbe07fc7c6071db6db257.png"},{"id":94470000,"identity":"e2766ab6-ae91-4d03-a667-9f1ad3722abc","added_by":"auto","created_at":"2025-10-27 15:31:22","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":83939,"visible":true,"origin":"","legend":"\u003cp\u003eDendrogram structure generated from cluster analysis.\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/f8df231bf6f03c0e29554595.png"},{"id":94470002,"identity":"75f79f99-0eeb-4714-8374-0f44ac976443","added_by":"auto","created_at":"2025-10-27 15:31:22","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":138614,"visible":true,"origin":"","legend":"\u003cp\u003eChart showing the feature importance of each parameter.\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/b9aec88151159f667d80e3c8.png"},{"id":94469777,"identity":"8d71e89a-498e-4f5c-8eb5-7364e3ec5106","added_by":"auto","created_at":"2025-10-27 15:30:38","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":234056,"visible":true,"origin":"","legend":"\u003cp\u003eChart showing the proportional contribution of each region by parameter\u003c/p\u003e","description":"","filename":"floatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/afeef7447143e49591f40897.png"},{"id":100070122,"identity":"8783d953-0c2b-4d4b-b586-428b14548aef","added_by":"auto","created_at":"2026-01-12 16:16:33","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4278126,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7819579/v1/bfd3e76c-0eac-4f95-978b-6056a240050c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eSoil Geochemistry and Contamination Zoning in Northeastern Ghana: Insights From the Bongo and Talensi Districts\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eSoil geochemical studies require knowledge of natural chemical element concentrations to differentiate between anthropogenic pollution from lithogenic signatures in environmental frameworks. This foundational concept, termed geochemical background, has its origins in exploration geochemistry, where it was traditionally used to delineate mineral anomalies from natural levels of chemical constituents in geological units (Hawkes \u0026amp; Webb, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e1962\u003c/span\u003e; Abimbola \u0026amp; Olatunji, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Environmental geochemistry now incorporates geochemical background as a core principle which enables researchers to study both natural system variations and human-induced impacts on Earth surface chemicals (Filzmoser et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). Geochemical background is defined as the normal concentration range of elements in environmental media that are unaffected by human activity. The term geochemical baseline which gained popularity through International Geological Correlation Programme initiatives depicts a momentary element concentration measurement but may contain anthropogenic component stain from sources (Salminen \u0026amp; Tarvainen, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). The baseline provides understanding of both natural background levels and weak effects from human activities and airborne pollutants while presenting improved temporal analysis opportunities according to Johnson et al. (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2005\u003c/span\u003e) and Reimann et al. (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eDespite its recognized significance, studies aimed at systematically determining geochemical background and baseline values remain scarce in Ghana and, more broadly, across West Africa (Amuah et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Kazapoe and Arhin, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This research gap is particularly evident in northern Ghana, where regions such as the Bongo and Talensi districts of the Upper East Region exhibit complex geological settings with limited geochemical characterization. Talensi is particularly heavily influenced by artisanal and small-scale gold mining (ASGM), which poses a significant risk of soil contamination and geochemical distortion. Soil contamination assessment becomes complicated when there is no access to verified baseline and background data that help determine specific contamination thresholds needed for proper cleanup efforts (Armah et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Bempah \u0026amp; Ewusi, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe geological framework of Bongo and Talensi is predominantly composed of Paleoproterozoic Birimian rocks and associated granitoids, which are renowned for their metallogenic significance, especially with regard to gold mineralization. Because of various geological formations with related weathered features these districts present complications that may probably interfere with the distribution patterns of elements (e.g. Bayari et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) thus demanding separation between weathered geologic signatures and human-caused environmental alterations.\u003c/p\u003e\u003cp\u003eThis research addresses the critical data shortage regarding basic geochemical values by evaluating major and trace elements in district soils to identify natural background thresholds. This study therefore combines traditional methods with the multifractal method to determine regional geochemical background values and baseline values in the Talensi and Bongo districts of the Upper East region of Ghana. Machine learning is further used to comprehensively evaluate the geochemical indicator importance through which it delineates geochemical zones across various study areas. This approach enhances our understanding of the spatial patterns and variability of geochemical components across the study area. It provides a scientifically sound foundation for the exploration, evaluation, and development of natural resources.\u003c/p\u003e\u003cp\u003eBy using this framework for the first time in northern Ghana, essential findings about both geochemical spatial distribution across the area and pollution risks from mining activities can be established. The researched outcomes will add substantial value to academic publications combined with local environmental governance because they can guide future approaches to assessing pollution and land use planning and natural resource management.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Study Area\u003c/h2\u003e\u003cp\u003eThe study was conducted in the Bongo and Talensi districts, located in the Upper East Region of Ghana (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This geographical area belongs to the savannah ecological zone where semi-arid climate prevails. The Bongo District occupies a geographic position between 10\u0026deg;45\u0026prime;N and 11\u0026deg;10\u0026prime;N and 0\u0026deg;48\u0026prime;W and 1\u0026deg;05\u0026prime;W while sharing western borders with Kassena-Nankana and southern boundaries with Bolgatanga Municipality and northern and eastern edges bordering Burkina Faso (Ministry of Food and Agriculture [MoFA], 2024). The Talensi District extends between 0\u0026deg;30\u0026prime;00\u0026Prime;W to 0\u0026deg;45\u0026prime;00\u0026Prime;W west longitude and 10\u0026deg;35\u0026prime;00\u0026Prime;N to 10\u0026deg;50\u0026prime;00\u0026Prime;N north latitude. The area borders Bolgatanga District to its north whereas Kassena-Nankana District occupies its west and Bawku West District stands to its east while West and East Mamprusi districts position themselves to the south (MoFA, 2024).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe geology of both areas shows Palaeoproterozoic lithologies belonging to the Birimian Supergroup (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The basement structure of the West African Craton consists of metavolcanic and metasedimentary rocks along with granitoid intrusions according to Griffis et al. (2002) and Kazapoe (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These formations are known to host gold-bearing quartz veins, making the region historically significant for both mechanized and artisanal gold mining activities (Kwayisi et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe districts show a smooth topographical layout because their slopes range from 1% to 5%. The locations at Tongo and Nangodi and their adjacent steep terrain have slopes reaching up to 10% that trigger occasional soil losses through mass movements and erosion. Granitic and volcanic rock outcrops disrupt the area while influencing soil development and drainage system formation (MoFA, 2024). The area experiences seasonal water drainage from White and Red Volta Rivers through their tributaries. The tributary networks offer limited water resources during rainy months which helps charge underground aquifers that serve as the main water source for both domestic uses and farming operations.\u003c/p\u003e\u003cp\u003eBoth Bongo and Talensi districts have their economies based primarily on small-scale agricultural farming with additional artisanal mining activities. Among the dispersed rural communities of Bongo District most residents cultivate millet and sorghum. Similarly, the Talensi District, with a population density of about 99 persons per square kilometre, combines dryland agriculture with intensive artisanal mining, particularly in and around Nangodi (Ghana Statistical Service [GSS], 2021; MoFA, 2024). The use of the land under these practices creates pressure which degrades soils and exposes humans to dangerous elements.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Data Preparation\u003c/h2\u003e\u003cp\u003eGeochemical datasets were compiled for two study areas Bongo and Talensi, in North-Eastern Ghana (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The datasets consist of concentrations for a suite of major oxides and trace elements derived from previously collected soil samples. The datasets were cleaned and appropriate conversions undertaken.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Determination of Geochemical Background Values\u003c/h2\u003e\u003cp\u003eThree methods were employed for estimating the geochemical background values of examined elements and compounds. Researchers executed all implementation of these methods through Python 3.10 by employing the \u0026ldquo;scipy.stats\u0026rdquo; library to perform statistics along with \u0026ldquo;matplotlib/seaborn\u0026rdquo; library to create visualizations.\u003c/p\u003e\u003cdiv id=\"Sec6\" class=\"Section3\"\u003e\u003ch2\u003e2.3.1 Method 1: Iterative Method\u003c/h2\u003e\u003cp\u003eAccording to Wen et al (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) the iterative method served as the calculations basis to determine geochemical background values (GBVs) for study elements in this work. Systematic removal of extreme values helps the iterative method transform non-normal distributions into ones which more closely resemble normal distributions. A computation of mean (X₁) and standard deviation (SD₁) occurs for each element. A set of potential outliers gets identified during the first iteration through a values comparison against X₁ \u0026plusmn; 3SD₁ range. The second set of calculations for X₂ and SD₂ uses all remaining values obtained after removing the initial outlier data for the second iteration. The procedure continues toward successive iterations to determine additional outliers until the third computation (X₃ and SD₃) with modified outlier elimination. The process will repeat infinitely until all outliers disappear from the loop (Xₙ \u0026plusmn; 3SDₙ).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section3\"\u003e\u003ch2\u003e2.3.2 Method 2: Frequency Distribution\u003c/h2\u003e\u003cp\u003eThe frequency distribution method employed for calculating geochemical background values was also adapted from (Wen et al, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Before implementation, the dataset was first subjected to log(x\u0026thinsp;+\u0026thinsp;1) transformation to reduce right-skewness and normalize distributions with zero or near-zero values. Frequency histograms were then constructed for the transformed data. These histograms helped visualize the modal classes, this is the class with the highest frequency, the spread of values, and any anomalies at the tails. To objectively determine the crowd value (M₀), the method involves projecting two geometric connections across the histogram, firstly, a line from the top-left corner of the modal class to the top-left corner of the next class to the right. Next, another line from the top-right corner of the modal class to the top-right corner of the preceding class to the left is drawn. The point where these two lines intersect and project to the x-axis corresponds to M₀, considered the peak of the distribution. After this, a horizontal line is drawn at 60% of the maximum frequency (0.6f). The intersections of this line with both sides of the frequency curve marks the spread of data around M₀. The distance from M₀ to each of these intersection points defines the mean square error (σ). This σ is then used to establish the lower threshold of geochemical anomalies (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{M}_{0}+2\\sigma\\:)\\)\u003c/span\u003e\u003c/span\u003e. Mathematically, M₀ is calculated using:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:{M}_{0}=\\:{x}_{0}+\\frac{i({p}_{2}-{p}_{1})}{2{p}_{2}-{p}_{1}-{p}_{3}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{x}_{0}\\)\u003c/span\u003e\u003c/span\u003e is the lower boundary of the modal class, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:i\\)\u003c/span\u003e\u003c/span\u003e the width of class intervals, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{p}_{1}\\)\u003c/span\u003e\u003c/span\u003e the frequency of the class before the modal class, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{p}_{2}\\)\u003c/span\u003e\u003c/span\u003e the frequency of the modal class, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{p}_{3}\\)\u003c/span\u003e\u003c/span\u003e the frequency of the class after the modal class.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\u003ch2\u003e2.3.3 Method 3: (C\u0026ndash;A) Method\u003c/h2\u003e\u003cp\u003eIn this study, the C\u0026ndash;A method adapted from (Wen et al, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) was operationalized by constructing log\u0026ndash;log cumulative frequency plots of elemental concentrations. The cumulative frequency of the data points exceeding a given concentration threshold was plotted against those concentration values on a log\u0026ndash;log scale, yielding a piecewise linear trend that reflects the power-law behaviour of the dataset. For each element, concentrations were averaged from multiple samples then sorted in descending order. The cumulative frequency (rank/total) was computed for each value, and both the concentration and the cumulative frequency were transformed using base-10 logarithm. A double logarithmic plot, this is, log concentration to log cumulative frequency was constructed. In most cases, the data followed a multi-segment linear pattern, interpreted as follows. The first segment which is a steep slope typically reflects anomalously high values, often associated with mineralization or contamination. The second segment which is a gentler slope is assumed to represent the non-singular background population of the element in the environment. A linear regression model is then fitted to the second segment of the curve, and the arithmetic mean of the concentrations within this segment is calculated. This value is then recorded as the geochemical background value C₀₃ for that element. Unlike the frequency or iterative methods, this method does not rely on standard deviation thresholds to remove high values. Instead, it identifies thresholds through fractal breaks in slope on the log\u0026ndash;log plot, thereby improving background estimation in non-Gaussian datasets.\u003c/p\u003e\u003cp\u003eHaving determined geochemical background values using all three methods, the final geochemical background value, C₀ for each element was then obtained by averaging the values calculated from the three methods this is the Averaged Geochemical Background Values (AGBV).\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Determination of Geochemical Baseline Values (GBV) and Normalization\u003c/h2\u003e\u003cp\u003eThe baseline values, however, were determined using a distribution driven classification method and applied to the original dataset. This process begins with testing the data for normality, to classify the distribution of each element across the dataset. If the dataset conforms to a normal distribution, the arithmetic mean plus or minus two standard deviations is used to define the baseline value and its range. If the dataset adheres to a log-normal distribution, then the geometric mean is multiplied and divided by the square of the geometric standard deviation (GSD\u0026sup2;) to determine the baseline range. A geometric formulation maintains statistical consistency between skewed distributions that do not follow normal or log-normal patterns by taking asymmetry into account. The GBVs received their standardization treatment through UCC values adopted from Wedepohl (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e1995\u003c/span\u003e) and Taylor (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1964\u003c/span\u003e). The GBV/UCC and Background-to-Baseline Ratio measurements helped determine elemental excess levels compared to natural rock composition for detecting human sources of contamination. Factors exceeding one value suggested that the elements exceeded their natural distribution in the environment.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Cluster Analysis\u003c/h2\u003e\u003cp\u003eThe hierarchical cluster analysis (HCA) assessed elemental distribution patterns by utilizing background values of trace elements collected from Bongo and Talensi. The unguided learning approach examined element distribution patterns to determine behavioural clusters between geochemical entities and locate distinctive compositional signatures of each location. The analysis selected trace elements exclusively to prevent major oxide scale bias that occurs because these elements exist at much higher concentrations. Z-score normalization standardization techniques applied to data allowed element comparisons despite various unit differences and magnitude levels. The dataset transformation produced a unitless scale that positioned each variable at zero standard deviation while maintaining a measurement scale of one which created equal importance for all trace elements during clustering. Euclidean distance measures the dissimilarities between clusters while Ward\u0026rsquo;s linkage method executes the clustering process. A Heatmap Visualization of Dendrogram displayed the HCA output.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Machine Learning for Feature Importance\u003c/h2\u003e\u003cp\u003eA supervised machine learning technique served to evaluate additional aspects about geochemical indicator differences between the Bongo and Talensi regions. The method enabled the determination of the relative prominence of each major and trace element in establishing regional classifications while strengthening geochemical zoning strategies. The input dataset consisted of the original dataset for both regions, with elements and compounds serving as independent variables (features) and the region (Bongo or Talensi) as the dependent variable (target). The regions were one-hot encoded to facilitate regression-based modelling. All features were standardized using Z-score normalization* to ensure uniform influence across variables of differing scales and units.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e2.7 Model implementation\u003c/h2\u003e\u003cp\u003eA regression model was employed as the primary predictive algorithm. To increase its robustness, Lasso (L1) and Ridge (L2) regularization were implemented through Elastic Net models were tested for comparison. This helps minimize overfitting, a common risk when working with high-dimensional geochemical data and a limited number of classes. Model performance was evaluated using R\u0026sup2; and Adjusted R\u0026sup2; for explained variance, Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) for prediction accuracy. Cross-validation (k-fold) was also implemented to ensure model generalizability. Lastly, a hyperparameter tuner was employed using GridSearchCV to identify the optimal set of parameters for the model variant. The absolute values of the regression coefficients from the best-performing model were extracted to estimate the relative feature importance of each element. Higher absolute coefficient values indicated a greater role of that element in distinguishing between the two regions.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Geochemical Background Values\u003c/h2\u003e\u003cp\u003eMost of the elemental indicators achieved internal consistency from background values estimated through Iterative 3σ, Frequency Histogram and Concentration-Area (C\u0026ndash;A) methods. Geochemical distribution patterns throughout the Bongo and Talensi regions demonstrate a stable central value presence yet show medium-level effects from distributional skewness in addition to outlier-induced variations, particularly for Cd, Sb and Ag trace elements. The minor differences in the method stem from differences in tail effects sensitivity and distribution assumption sensitivity yet match Reimann and Garrett\u0026rsquo;s (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2005\u003c/span\u003e) view that no single method operates on all geochemical data types since natural systems lack normality.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Baseline Values\u003c/h2\u003e\u003cp\u003eThe baseline values were calculated by using the arithmetic mean\u0026thinsp;\u0026plusmn;\u0026thinsp;2SD for normal distribution elements and the geometric mean for elements following a log-normal distribution. The baseline ranges for elements Cd and Ge required adjustment because these elements show substantial skewness and have naturally high concentration patterns across lithologic units. The separation of natural and anthropogenic activities is essential because hazardous mining effects may amplify the existing geological variations in Talensi. The baseline calculations (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) indicated that elemental levels in most tested samples maintained typical natural concentrations, demonstrating that the studied areas' geochemical status mostly remained pristine until elements showed increased patterns in specific loci.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e3.3 UCC and Baseline Ratios\u003c/h2\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the background to UCC enrichment factors for the Bongo district. Comparison with Upper Continental Crust (UCC) values (Wedepohl, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e1995\u003c/span\u003e) highlighted pronounced regional differences in elemental enrichment. In the Talensi domain (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e), the ratios of Cd, Sb and Ag surpassed unity, whereas other elements remained below unity, which could result from natural or human disturbances to the area. The measured levels of elements likely result from aspects of sulfide mineralisation properties, together with natural baseline variations and former mining activities across West African Birimian regions (BGS, 1996).\u003c/p\u003e\u003cp\u003eThe geochemical analysis shows SiO₂, K₂O, and Al₂O₃ stay under UCC values or align within UCC constraints in Bongo, particularly because of its felsic and weathered geologic origin with negligible metallogenic enhancements. Geological studies confirm that the Bongo area contains granitic gneiss rocks, which contrast with Talensi (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) where metavolcanic belts hold the potential for Au-As mineralisation.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Geochemical Grouping using Ternary Diagrams\u003c/h2\u003e\u003cp\u003eThe three ternary diagrams of (SiO₂/10\u0026ndash;Al₂O₃\u0026ndash;CaO), (K₂O\u0026thinsp;+\u0026thinsp;Na₂O\u0026ndash;CaO\u0026ndash;TFe₂O₃) and (CaO\u0026ndash;K₂O\u0026ndash;Na₂O) from Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e respectively, show separate groupings which demonstrate the distinct mineralogical and geochemical settings of Bongo and Talensi regions. Bongo rocks accumulate in the silica-rich end-member on the SiO₂/10\u0026ndash;Al₂O₃\u0026ndash;CaO diagram (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e) because they contain quartz and show weathering features. Talensi samples show higher proportions of CaO and Fe₂O₃ components when comparing equivalent elemental proportions because their mineral compositions are heavily influenced by ferromagnesian silicates as well as carbonates.\u003c/p\u003e\u003cp\u003eTalensi samples in the (K₂O\u0026thinsp;+\u0026thinsp;Na₂O)\u0026ndash;CaO\u0026ndash;TFe₂O₃ diagram (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e) exhibit higher TFe₂O₃ proportions that match mafic source rock weathering and mineralised formations. Soil chemistry across the region exhibits these geochemical associations because they result from the interaction between lithological variations, mineral weathering mechanisms and metal transport patterns in the area (Reimann et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Cluster Analysis\u003c/h2\u003e\u003cp\u003eThe dendrogram structure generated from cluster analysis divides the elemental data into two major domains corresponding to the geographical extent of Bongo and Talensi (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e). The group of trace elements found in the Talensi cluster shows a uniform geochemical signature that comes from common mineral sources or substance transport routes which link to ore deposits and human-made soil disturbance. The Bongo cluster consists of lithophile elements that initially came from felsic and less metalliferous rock.\u003c/p\u003e\u003cp\u003eFurther exploration is justified at sub-regional levels because geochemical region-specific zoning patterns have been identified while also benefiting environmental monitoring and mineral exploration needs.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003e3.6 Machine Learning for Feature Importance\u003c/h2\u003e\u003cdiv id=\"Sec20\" class=\"Section3\"\u003e\u003ch2\u003e3.6.1 Model Evaluation\u003c/h2\u003e\u003cp\u003eThe model performance metrics indicated a moderately strong fit, suggesting that geochemical background values hold substantial predictive power for regional classification. The results suggest that approximately 66% of the variance in regional differentiation could be explained by the model, with prediction errors kept within a relatively narrow margin.\u003c/p\u003e\u003cp\u003eThe regression model identified several geochemical variables as key contributors to regional differentiation between Bongo and Talensi. Feature importance (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e) was assessed based on the absolute value of the standardised regression coefficients, with higher values indicating greater influence on the classification outcome. Performance metrics acquired from using the regression model demonstrate a moderate level of fit with R\u0026sup2; = 0.6623 and Adjusted R\u0026sup2; = 0.5675 in addition to RMSE\u0026thinsp;=\u0026thinsp;0.2778 and MAE\u0026thinsp;=\u0026thinsp;0.2029. The elemental predictors account for sixty-two percent of the variation that exists in regional geochemical differentiation patterns.\u003c/p\u003e\u003cp\u003eUsing standardised regression coefficients, we determined the feature importance where Cd emerged as the primary variable and Sb and Ge followed behind, while Ag ranked fourth. Multiple trace metals and metalloids rank foremost among top contributors, which signifies that region-type determination relies mostly on elements which naturally associate with mineralisation and show heightened sensitivity to earth and anthropogenic sources. Research in identical African locations has proven that heavy metals operate as effective markers to track both geological wealth potential and human-caused environmental damage (Obiri et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe regression model proved effective, yet it requires recognition of its insufficient capability to handle specific situations. Other non-linear influencing factors that influence elemental concentrations include redox conditions and micro-topography, and mineral grain size because they are not fully represented in the prediction model.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003e3.7 Discussions\u003c/h2\u003e\u003cp\u003eCollectively, the geochemical, statistical, and machine learning results converge on a robust delineation of geochemical provinces within the Upper East Region of Ghana. The Talensi area displays increased trace metals as well as distinct geochemical clusters combined with volcanic and possibly hydrothermal indicators in its mineralogy. The Bongo region displays geochemical indicators consisting mainly of silicate weathering materials in addition to minimal trace elements and limited alteration of chemical signatures.\u003c/p\u003e\u003cp\u003eThe analysis demonstrated that Cr appeared as the dominant trace element, reaching 276.57 mg/kg in the final average. Zr and strontium Sr followed, with concentrations of 1168.17 mg/kg and 371.88 mg/kg, respectively. The V levels in the samples were 29.07 mg/kg but Cu and Pb levels were lower at 7.87 mg/kg and 4.82 mg/kg, respectively. The elevated Cr concentrations suggest a significant contribution from mafic to ultramafic bedrock sources, corroborating interpretations of a lithological influence on soil chemistry (Taylor \u0026amp; McLennan, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1995\u003c/span\u003e). The coherence among the three baseline methods was generally high, although V showed substantial variation, as its results varied from 13.01 to 39.64 mg/kg. Such variation is expected given that the (C\u0026ndash;A) plot method, in particular, tends to highlight localised enrichments associated with mineralised zones as described in Cheng (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe studied concentrations of Cu and Pb, and Zn remained under the established international standards for critical thresholds. The study showed Bongo soils contained less Cu than two worldwide benchmarks \u0026ndash; the Canadian Council of Ministers of the Environment guideline, which amounts to 63 mg/kg, while the United States Environmental Protection Agency residential limit stands at 1600 mg/kg (CCME, 2007; USEPA, 2018). Similarly, Zn concentrations remained far below the 200 mg/kg CCME guideline. The Cr concentration in the Bongo soils exceeded both the CCME threshold of 64 mg/kg and the VROM threshold of 100 mg/kg, which reveals a geogenic natural enrichment process instead of human-induced pollution. This enrichment is consistent with globally reported behaviour of Cr in tropical lateritic profiles developed over ultramafic bedrock (Kabata-Pendias, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe analysis revealed that Talensi showed higher Mn levels at 77.04 mg/kg in addition to Zn levels at 26.15 mg/kg as compared to Bongo (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e). The measured Cr content was 185.12 mg/kg in Talensi soils and exceeded the CCME and VROM soil quality standards yet remained lower than Bongo values. Notably, Co averaged 19.56 mg/kg, while Ni concentrations showed high variability, with values ranging between 0.25 and 17.59 mg/kg depending on the method used. The high variability in Ni measurements indicates that either particular concentrations exist in specific locations or that chemical substance variability in the native parent materials happens across the area.\u003c/p\u003e\u003cp\u003eThe concentrations of Cu and Pb were reported as 22.06 mg/kg and 4.76 mg/kg, and As concentrations averaged 0.82 mg/kg throughout the Talensi area, where all values persisted under the USEPA residential soil guidelines. When benchmarked against international standards, Bongo District soils exhibited mixed trends. The high Cr levels in the soil match both CCME and VROM soil quality limits, thus proving the lithogenic enrichment theory. Soils contained substantial amounts of vanadium, although their levels fell below the World average values at 42.5 mg/kg while remaining at normal background ranges. The Ba content in the Bongo District soils reached 1038.37 mg/kg beyond the Tanzanian Bureau of Standards (TBS) guideline of 100 mg/kg but is similar to standard values for felsic rock-derived soils (Kabata-Pendias, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). The soil's analytical results showed that Pb, Cu, Zn, Co and As levels remained under both Tanzanian Bureau of Standards limits and international intervention thresholds, while Pb concentrations exceeded TBS guidelines.\u003c/p\u003e\u003cp\u003eThe Cr content in Talensi soil matched the distribution patterns found in Bongo within international standards. Additionally, the Cr measurements stood above both CCME (64 mg/kg) and VROM (100 mg/kg) thresholds. Soil measurements of V reached or slightly exceeded the World natural background standards. The analysed Ba content reached 465.25 mg/kg, which exceeded the TBS threshold yet maintained values found in mineral-rich soils according to Kabata-Pendias (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). The environmental health risks were minimal because Zn, Pb, Cu, Co and As content did not surpass allowable thresholds.\u003c/p\u003e\u003cp\u003eThe geochemical characteristics of Bongo show predominantly geogenic influences on element distribution patterns, mainly for Cr and Ba. The detected elements' levels indicated no major human-made contamination because they remained low at various sampling sites. Intense lateritization zones exist in the region based on moderate to elevated concentrations of Al₂O₃ and Fe₂O₃, which enhance the potential for finding mineralised structures below the surface. The geographically distributed Cr and Ba patterns reveal separate rock formations, which will help improve upcoming environmental examinations and mineral resource evaluations for the district (Dill, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eVarying levels of trace elements exist more broadly throughout Talensi territorial space as compared to Bongo. Higher concentration variability of V, Ni and Zn demonstrates that the soil chemistry is influenced by more heterogeneous bedrock in addition to surficial processes. The elevated Cr and V levels within the area most likely originate from specific units of the greenstone belt in addition to weathered ultramafic rock formations. The heterogeneous geochemical conditions require site-specific baseline assessments instead of generalized regional approaches (Taylor \u0026amp; McLennan, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1995\u003c/span\u003e) because they impact environmental risk evaluations and mineral exploration activities.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThe geochemical baselines established for the Bongo and Talensi districts of northeastern Ghana illustrate a predominantly lithogenic control over soil chemistry, particularly concerning Cr and Ba concentrations. While Cu, Pb, Zn, Co, and As levels remain low, natural enrichments in chromium and barium necessitate careful consideration in land use planning and environmental assessments. Advanced combination methods between Iterative and Frequency Distribution and Concentration\u0026ndash;Area have produced reliable baseline data that serves as guidance for upcoming environmental management and mapping operations. This research establishes that natural geochemical variations control the soil composition of these districts more than human-made contaminations do.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBelinda S. Berdie conceptualized the study, led the fieldwork and data analysis and co-authored the manuscript. Raymond W. Kazapoe handled geostatistical modelling and regional geological interpretation and co-authored the manuscript. Darwin A. Awog-Badek handled the machine learning implementation and visualisation. Blestmond A. Brako, Gordon Foli and Simon Y. Gawu supported literature review, provided supervisory guidance and critical manuscript refinement. All authors reviewed the final manuscript and approved its submission.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by the Ghana Education Trust Fund (GETFUND), and the Ghana National Petroleum Commission (GNPC) Foundation. Institutional support, including laboratory facilities and supervision was provided by C. K. Tedam University of Technology and Applied Sciences (CKT-UTAS), Kwame Nkrumah University of Science and Technology (KNUST), University of Ghana (UG) and the Ghana Geological Survey Authority (GGSA). The authors gratefully acknowledge these organisations for their role in facilitating the successful completion of this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData will be made available upon reasonable request from the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbimbola, A. F., \u0026amp; Olatunji, A. S. (2011). Urban geochemical mapping in Nigeria with some examples from southern Nigeria. \u003cem\u003eMapping the Chemical Environment of Urban Areas\u003c/em\u003e, 570-580.\u003c/li\u003e\n\u003cli\u003eAmuah, E.E.Y., Fei-Baffoe, B., Kazapoe, R.W., Dankwa, P., Okyere, I.K., Sackey, L.N.A., Nang, D.B. and Kpiebaya, P. (2024). From the ground up: Unveiling Ghana\u0026apos;s soil quality crisis and its ecological and health implications. \u003cem\u003eInnovation and Green Development\u003c/em\u003e, 3(1), p.100097.\u003c/li\u003e\n\u003cli\u003eArmah, F. A., Quansah, R., \u0026amp; Luginaah, I. (2014). A systematic review of heavy metals of anthropogenic origin in environmental media and biota in the context of gold mining in Ghana. \u003cem\u003eInternational scholarly research notices\u003c/em\u003e, 2014(1), 252148. https://doi.org/10.1155/2014/252148 \u003c/li\u003e\n\u003cli\u003eBayari, E. E., Foli, G. and Gawu, S. K. Y. (2019). Geochemical and pathfinder elements assessment in some mineralised Regolith profiles in Bole-Nangodi gold belt in north-eastern Ghana. \u003cem\u003eEnvironmental Earth Sciences\u003c/em\u003e, 78(9), 1-16. https://doi.org/10.1007/s12665-019-8265-5 \u003c/li\u003e\n\u003cli\u003eBempah, C. K., \u0026amp; Ewusi, A. (2016). Heavy metals contamination and human health risk assessment around Obuasi gold mine in Ghana. \u003cem\u003eEnvironmental monitoring and assessment\u003c/em\u003e, 188, 1-13. https://doi.org/10.1007/s10661-016-5241-3\u003c/li\u003e\n\u003cli\u003eBritish Geological Survey (BGS). (1996). \u003cem\u003eGeochemical Atlas of Ghana\u003c/em\u003e. Keyworth, Nottingham, Uk..\u003c/li\u003e\n\u003cli\u003eCanadian Council of Ministers of the Environment (CCME). (2007). \u003cem\u003eCanadian soil quality guidelines for the protection of environmental and human health. \u003c/em\u003eRetrieved from https://ccme.ca \u003c/li\u003e\n\u003cli\u003eCheng, Q. (2007). Mapping singularities with stream sediment geochemical data for mineral exploration. \u003cem\u003eOre Geology Reviews\u003c/em\u003e, 32(1\u0026ndash;2), 314\u0026ndash;324.\u003c/li\u003e\n\u003cli\u003eDill, H. G. (2010). The geology of aluminum phosphates and sulfates of the alunite group minerals: An overview. \u003cem\u003eEarth-Science Reviews\u003c/em\u003e, 101(1\u0026ndash;2), 1\u0026ndash;35.\u003c/li\u003e\n\u003cli\u003eFilzmoser, P., Garrett, R. G., \u0026amp; Reimann, C. (2005). \u003cem\u003eMultivariate outlier detection in exploration geochemistry\u003c/em\u003e. Computers \u0026amp; geosciences, 31(5), 579-587.\u003c/li\u003e\n\u003cli\u003eGhana Statistical Service (GSS). (2021). 2021 Population and Housing Census: population of Regions and Districts (Vol. 3A). Accra, Ghana. https://statsghana.gov.gh/gssmain/fileUpload/pressrelease/2021%20PHC%\n20General%20Report%20Vol%\n203A_Population%20of%20Re\ngions%20and%20Districts_181121.pdf\u003c/li\u003e\n\u003cli\u003eHawkes, H.E. and Webb, J.S., 1962. \u003cem\u003eGeochemistry in Mineral Exploration\u003c/em\u003e, first edition, Harper \u0026amp; Row, New York.\u003c/li\u003e\n\u003cli\u003eJohnson, C. C., Breward, N., Ander, E. L., \u0026amp; Ault, L. (2005). G-BASE: baseline geochemical mapping of Great Britain and Northern Ireland. \u003cem\u003eGeochemistry: exploration, environment, analysis\u003c/em\u003e, 5(4), 347-357.\u003c/li\u003e\n\u003cli\u003eKabata-Pendias, A. (2011). \u003cem\u003eTrace elements in soils and plants\u003c/em\u003e (4th ed.). CRC Press.\u003c/li\u003e\n\u003cli\u003eKazapoe, R. W. (2023). A review of the characteristics and geological settings of orogenic gold deposits of the Boule Mossi Domain: Implication for gold exploration. \u003cem\u003eGeology, Ecology, and Landscapes\u003c/em\u003e, 9(2), 690-705. https://doi.org/10.1080/24749508.2023.2256553 \u003c/li\u003e\n\u003cli\u003eKazapoe, R. W. \u0026amp; Arhin, E. (2021). Determination of local background and baseline values of elements within the soils of the Birimian Terrain of the Wassa Area of Southwest Ghana. \u003cem\u003eGeology, Ecology, and Landscapes\u003c/em\u003e, 5(3), 199-208.\u003c/li\u003e\n\u003cli\u003eKazapoe, R. W., \u0026amp; Arhin, E. (2019). Selenium in locally produced food crops and implications on healthy eating: A case study at the Talensi District of Ghana\u003cem\u003e. EC Nutrition\u003c/em\u003e, 8(3), 85\u0026ndash;92.\u003c/li\u003e\n\u003cli\u003eKesse, G. O. (1985). \u003cem\u003eThe mineral and rock resources of Ghana\u003c/em\u003e. A.A. Balkema.\u003c/li\u003e\n\u003cli\u003eKwayisi, D., Kazapoe, R. W., Alidu, S., Sagoe, S. D., Umaru, A. O., Amuah, E. E. Y. \u0026amp; Kpiebaya, P. (2024). Exploring soil pollution patterns in Ghana\u0026rsquo;s northeastern mining zone using machine learning models. \u003cem\u003eJournal of Hazardous Materials Advances\u003c/em\u003e, 16, 100480.\u003c/li\u003e\n\u003cli\u003eMatschullat, J., Ottenstein, R., \u0026amp; Reimann, C. (2000). Geochemical background\u0026mdash;can we calculate it? \u003cem\u003eEnvironmental Geology\u003c/em\u003e, 39(9), 990\u0026ndash;1000.\u003c/li\u003e\n\u003cli\u003eMcLaughlin, M. J., Parker, D. R., \u0026amp; Clarke, J. M. (2000). Metals and micronutrients \u0026ndash; food safety issues. \u003cem\u003eField Crops Research\u003c/em\u003e, 60(1\u0026ndash;2), 143\u0026ndash;163.\u003c/li\u003e\n\u003cli\u003eMinistry of Food and Agriculture. (2024). \u003cem\u003eUpper East Region Agricultural Sector Projects and Potentials\u003c/em\u003e. Retrieved from https://mofa.gov.gh/site/directorates/50-regional-directorates/upper-east-region \u003c/li\u003e\n\u003cli\u003eObiri, S., Dodoo, D. K., \u0026amp; Okai-Sam, F. (2016). Cancer and non-cancer health risk from exposure to arsenic, cadmium, and copper in borehole, tap, and surface water in the Obuasi municipality of Ghana. \u003cem\u003eHuman and Ecological Risk Assessment\u003c/em\u003e, 22(2), 310\u0026ndash;323. https://doi.org/10.1080/10807031003788907 \u003c/li\u003e\n\u003cli\u003eReimann, C., \u0026amp; Garrett, R. G. (2005). Geochemical background\u0026mdash;concept and reality. \u003cem\u003eScience of the Total Environment\u003c/em\u003e, 350(1\u0026ndash;3), 12\u0026ndash;27.\u003c/li\u003e\n\u003cli\u003eReimann, C., Fabian, K., Flem, B., Andersson, M., Filzmoser, P., \u0026amp; Englmaier, P. (2018). Geosphere-biosphere circulation of chemical elements in soil and plant systems from a 100 km transect from southern central Norway. \u003cem\u003eScience of the Total Environment\u003c/em\u003e, 639, 129-145.\u003c/li\u003e\n\u003cli\u003eReimann, C., Filzmoser, P., \u0026amp; Garrett, R. G. (2008). \u003cem\u003eStatistical data analysis explained: Applied environmental statistics with R\u003c/em\u003e. John Wiley \u0026amp; Sons.\u003c/li\u003e\n\u003cli\u003eSalminen, R., \u0026amp; Tarvainen, T. (1997). The problem of defining geochemical baselines. A case study of selected elements and geological materials in Finland. \u003cem\u003eJournal of geochemical exploration\u003c/em\u003e, 60(1), 91-98.\u003c/li\u003e\n\u003cli\u003eTaylor, S. R., \u0026amp; McLennan, S. M. (1995). The geochemical evolution of the continental crust. \u003cem\u003eReviews of Geophysics\u003c/em\u003e, 33(2), 241\u0026ndash;265.\u003c/li\u003e\n\u003cli\u003eTaylor, S. R. (1964). Abundance of chemical elements in the continental crust: Anew table. \u003cem\u003eGeochimica et Cosmochimica Acta\u003c/em\u003e, 28(8), 1273-1285. https://doi.org/10.1016/0016-7037(64)90129-2 \u003c/li\u003e\n\u003cli\u003eUnited States Environmental Protection Agency (USEPA). (2018). \u003cem\u003eRegional Screening Level (RSLs) Generic Table\u003c/em\u003e. Retrieved from https://www.epa.gov/risk/regional-screening-levels-rsls-generic-tables \u003c/li\u003e\n\u003cli\u003eWedepohl, K. H. (1995). The composition of the continental crust. \u003cem\u003eGeochimica et Cosmochimica Acta\u003c/em\u003e, 59(7), 1217\u0026ndash;1232.\u003c/li\u003e\n\u003cli\u003eWen, W., Yang, F., Xie, S., Wang, C., Song, Y., Zhang, Y., \u0026amp; Zhou, W. (2024). Determination of Geochemical Background and Baseline and Research on Geochemical Zoning in the Desert and Sandy Areas of China. \u003cem\u003eApplied Sciences\u003c/em\u003e, 14(22), 10612.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"environmental-geochemistry-and-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"egah","sideBox":"Learn more about [Environmental Geochemistry and Health](https://www.springer.com/journal/10653)","snPcode":"10653","submissionUrl":"https://submission.nature.com/new-submission/10653/3","title":"Environmental Geochemistry and Health","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Soil Geochemistry, Geochemical background levels, multifractal modelling, machine learning, trace element distribution, northern Ghana","lastPublishedDoi":"10.21203/rs.3.rs-7819579/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7819579/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eUnderstanding natural geochemical background levels is essential for distinguishing anthropogenic pollution from lithogenic signatures in environmental studies. This research focuses on soil geochemical compositions in the Bongo and Talensi districts of northern Ghana, where limited geochemical characterization has hindered effective environmental assessments. This study applies advanced geostatistical methods including Iterative Frequency Distribution and Concentration-Area techniques, by integrating traditional geochemical analyses with multifractal modelling, to provide reliable baseline data for environmental management, land use planning, resource mapping. This study establishes regional geochemical background indicators and baseline values while employing machine learning techniques to evaluate geochemical indicators and delineate spatial geochemical zones. Results reveal distinct geochemical provinces, with Talensi showing increased trace metal concentrations and geochemical clusters linked to volcanic and possibly hydrothermal influences, while Bongo is characterized primarily by silicate weathering processes and minimal trace element variability. Chromium (Cr) emerges as the dominant trace element, with concentrations surpassing international threshold levels, indicating lithogenic enrichment rather than human-induced contamination. While Cu, Pb, Zn, Co, and As levels remain within regulatory standards, variations in Ni and V highlight the region\u0026rsquo;s heterogeneous bedrock influences. The findings demonstrate that natural geochemical variations \u0026ndash; rather than direct anthropogenic pollution \u0026ndash; primarily control soil chemistry in these districts. Concluding, the study offers crucial insights into geochemical spatial distribution while contributing to improved strategies for pollution assessment, mineral exploration and sustainable environmental governance in Ghana.\u003c/p\u003e","manuscriptTitle":"Soil Geochemistry and Contamination Zoning in Northeastern Ghana: Insights From the Bongo and Talensi Districts","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-27 13:55:56","doi":"10.21203/rs.3.rs-7819579/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-03T16:02:02+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-02T23:26:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"123211518315073268431840089066431896036","date":"2025-10-13T16:06:40+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-13T16:01:10+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-13T12:47:29+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-10T11:00:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"Environmental Geochemistry and Health","date":"2025-10-09T15:57:49+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"environmental-geochemistry-and-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"egah","sideBox":"Learn more about [Environmental Geochemistry and Health](https://www.springer.com/journal/10653)","snPcode":"10653","submissionUrl":"https://submission.nature.com/new-submission/10653/3","title":"Environmental Geochemistry and Health","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"3fa8f211-994f-49ff-b881-2e27775480e6","owner":[],"postedDate":"October 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-01-12T16:10:10+00:00","versionOfRecord":{"articleIdentity":"rs-7819579","link":"https://doi.org/10.1007/s10653-025-02967-y","journal":{"identity":"environmental-geochemistry-and-health","isVorOnly":false,"title":"Environmental Geochemistry and Health"},"publishedOn":"2026-01-08 15:59:10","publishedOnDateReadable":"January 8th, 2026"},"versionCreatedAt":"2025-10-27 13:55:56","video":"","vorDoi":"10.1007/s10653-025-02967-y","vorDoiUrl":"https://doi.org/10.1007/s10653-025-02967-y","workflowStages":[]},"version":"v1","identity":"rs-7819579","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7819579","identity":"rs-7819579","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}