Soil Pollution Mapping Across Africa: Potential Tool for Soil Health Monitoring

Abstract

There is an urgent need for an updated and relevant soil information system (SIS) to sustainably use and manage the land across Africa. Accurate data on soil pollution is essential for effective decision-making in soil health monitoring and management. Unfortunately, the data and information are not usually presented in formats that can easily guide decision-making. The objectives of this work were to (i) assess the availability of soil pollution maps, (ii) evaluate the methodologies used in creating these maps, (iii) explore the role of soil pollution maps in soil health monitoring, and (iv) identify gaps and challenges in soil pollution mapping in Africa. Soil pollution maps across Africa are created on a local scale, with highly variable sampling size and low sampling density. The most used mapping techniques include spatial interpolation (kriging and inverse distance weighting). Among the types of soil pollutants mapped, heavy metals have received priority, while pesticides and persistent organic pollutants have received less attention. Soil pollution mapping is not incorporated within the SIS framework due to lack of reliable spatially comprehensive data and technological and institutional barriers. Current efforts remain fragmented, site-specific, and methodologically inconsistent, resulting in significant data gaps that hinder reliable monitoring and limit progress in soil pollution mapping.

1. Introduction

The rapid growth of the human population has intensified pressures on soil and the broader environment, a challenge felt across Africa, leading to mismanagement of land and resulting in soil degradation [1,2]. According to the Intergovernmental Technical Panel on Soils (ITPSs) [3], one of the most serious threats to soils worldwide, alongside the ecosystem services they provide, is soil pollution. Soil pollution refers to the presence of a chemicals or substances that are out of place and/or present at a higher-than-normal concentration, causing adverse effects on any non-targeted organism [3]. Soil pollution is a global problem that has impacts at local and national levels, with transboundary effects. Soil is also the main recipient of environmental contaminants, and at the same time, it has the natural capacity to filter, buffer, retain, and degrade pollutants [4]. Soil pollutants are substances that contaminate or degrade soil quality, affecting its health, fertility, and the ability of plants, animals, and humans to thrive in that environment. In agricultural soils, pollutants can significantly affect soil quality, crop production, and the overall ecosystem. These pollutants commonly originate from human activities such as mining, chemical use, improper waste disposal, and unsustainable agricultural practices. In agricultural soils, pollution arises from the abusive use of pesticides and fertilizers, use of polluted irrigation water, and indiscriminate disposal of industrial waste. According to the FAO outcome document of the global symposium on soil pollution, the misuse of agricultural inputs, such as excessive fertilization and uncontrolled pesticide application, constitute the major sources of pollution in agricultural fields. A summary of major soil pollutants and their effects on agricultural soil and the environment at large [5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35] are shown in Table S1. In the context of the uncertain shocks caused by climate change, there is urgent need of up-to-date and relevant soil information that can be used for the sustainable management of soil resources [36]. Specifically, accurate data and information on soil pollution are essential for effective decision-making in soil health monitoring and management.

At the global level, many studies have been carried out to map soil pollution with the aim to understand the extent and impact of various pollutants on soil health and ecosystems. For example, Hou et al. [37] have established a global map for the distribution of heavy metals in agricultural soils, which indicates that 14 to 17% of croplands exceed agricultural thresholds for at least one toxic metal. In addition, Qi et al. [38] have established a global map for the distribution of heavy metal mobility and observe that about 37% of the world’s soils are at risk of medium to high levels of mobilization, with hotspots in Russia, Chile, Canada, and Namibia. Similar studies conducted at the global level exist for microplastic pollution [25,39,40] and excess salts (soil salinity) [41,42].

Europe and North America have established soil pollution monitoring initiatives, such as the Soil Health Dashboard [43] and EnviroAtlas, mapping contaminants such as pesticides, heavy metals, radionuclides, and microplastics [44,45,46,47,48,49], with similar efforts in Latin America, the Caribbean, China, and other parts of Asia [4]. In Africa, however, soil pollution data remains limited or difficult to access [4,50,51,52]. To address this, this study serves as an exploratory literature review, aimed at setting the stage for future and more extensive analyses to evaluate the availability and methodologies of soil pollution maps across African countries, examine their role in soil health monitoring, and identify existing gaps and challenges.

2. Major Soil Pollutants Across Africa and Their Effects

In Africa, soil pollution from heavy metals, pesticides, microplastics, and petroleum products, mainly from mining, industry, agriculture, and urbanization, poses serious environmental and health risks [53,54]. Weak regulations and limited awareness worsen contamination, as seen in Niger, where banned pesticides are misused [15]. Combined with the continent’s diverse climates and soils, these factors create regional variations in pollution, highlighting the urgent need to understand sources and impacts to guide effective soil management and safeguard human and ecosystem health.

2.1. Sources and Impacts of Heavy Metals

Heavy metal contaminants in Africa are prevalent in soil and water, posing significant risks to soil health and agricultural productivity. Cd, Hg, As, Zn, and Pb are some of the toxic elements deposited into soil during the mining and processing of minerals, causing serious harm to ecosystems and also to the health of human [55]. In Ethiopia, various activities such as industrial activities, artisanal mining, insufficient waste management systems, and agricultural practices have resulted in heavy metal pollution, causing many health challenges such as cardiovascular diseases, renal failure, cancer, and neurological disorders [55]. Apart from the risks to human health, mining activities also have serious environmental problems, such as soil and water contamination, loss of biodiversity, deforestation, and habitat destruction. Yevugah et al. [53] reported a high level of soil contamination with mercury in one-third of Ghana due to gold mining activities, which is three times higher than the level in the rest of the country. Dehkordi et al. [54] equally recorded soil contamination by heavy metals including Pb: 50–100 mg/kg, Zn: 25–50 mg/kg, and Cd: 5–10 mg/kg in Cairo, Egypt; and Au: 0.25–0.3 mg/kg, Pb and Zn: 1000–10,000 mg/kg in Kabwe, Zambia. Based on the permissible limits for heavy metals in soils, the concentrations of Pb and Zn reported in Zambia exceed the permissible limit of 300 mg/kg, placing them within the toxic range [45]. Similar pollution trends have been reported in many parts of Africa, especially where mining activities are intensely carried out, such as in South Africa [56]. In the southern part of Morocco, Boularbah et al. [57,58] made use of a rapid biotest kit to assess the concentrations of heavy metals in soils from three mining sites and found concentrations ranging from 0.2–31.5 mg/kg, 42–1683 mg/kg, 32–5756 mg/kg, and 173–8361 mg/kg for Cd, Cu, Pb, and Zn, respectively, with most of the concentrations surpassing the maximum permissible limits. The aforementioned studies indicated that the relatively high concentrations of these heavy metals were a major concern due to their potential toxicity to plants grown in such environments. By affecting soil quality, these pollutants pose a serious threat to agriculture and food security [52].

2.2. Soil Pollution and Impacts from Microplastics (MPs)

Microplastics (MPs) are an emerging global soil pollutant, and their use is widespread in African agriculture and industry due to their durability, portability, and cost-effectiveness [59]. Global plastic production reached approximately 368 million tons in 2019, with only 12% incinerated and 9% recycled [60]. In Africa, 25–33% of daily plastic waste is generated on the continent, posing serious risks to terrestrial and aquatic ecosystems [61]. Sources of MPs in soils include film mulching, sewage sludge application, wastewater irrigation, and atmospheric deposition [62]. Countries such as Nigeria, South Africa, and Egypt are the largest producers of plastic waste in Africa, with Egypt alone contributing 18.4%. Over time, plastic waste breaks down into MPs (≤5 mm), making soils significant reservoirs for their accumulation.

MPs are persistent pollutants that can transport heavy metals, pesticides, and antibiotics due to their hydrophobicity, small size, large surface area, and porous structure [63,64]. Studies have shown that MPs facilitate the uptake of metals such as Pb, Zn, and Cd by plants, increasing soil–plant toxicity [65]. Despite their importance, data on MP abundance in African soils remain limited [66]. Integrating MPs into soil pollution mapping efforts is therefore essential for identifying hotspots, assessing combined pollutant risks, and informing targeted management strategies, ensuring a comprehensive understanding of soil contamination across the continent.

2.3. Soil Pollution and Impacts from Indiscriminate Waste Dumps in African Agricultural Regions

Across Africa, many cities face the persistent challenge of indiscriminate or illegal waste dumping, which has significant implications for soil and groundwater quality [67,68,69]. Long-term disposal of domestic, municipal, and biodegradable wastes leads to leaching of harmful substances into soils, contaminating groundwater and posing risks to human health and ecosystems [68,69]. According to UNEP, Sub-Saharan Africa is a major destination for hazardous e-waste, with Nigeria and Ghana being among the highest recipients of these wastes in West Africa. Large quantities of this waste are also transported to the Republic of the Congo and to the Ivory Coast, while Tunisia, in North Africa, receives significant amounts of plastic waste from Europe [70]. Ineffective waste management, with only 44% of waste collected in the sub-region, exacerbates uncontrolled dumping [71]. The widespread and heterogeneous nature of these waste deposits creates highly variable and poorly documented patterns of soil contamination, presenting a major challenge for mapping soil pollution in agricultural and peri-urban areas and highlighting the urgent need for systematic spatial assessments.

2.4. Pesticide Pollution and Its Effects on Soil Quality in Africa

Using pesticides in agriculture is a pervasive practice across various African countries, aiming to control pests and enhance crop yields. However, the unregulated and excessive use of these chemicals has led to significant adverse effects on soil quality, compromising its overall health and productivity. Across Africa, studies aiming at investigating soil contamination by pesticides abound, with organochlorine pesticides dominating in soil environments [72]. These organochlorine pesticides occur in the following order: DDTs > HCHs >> endosulfans > dieldrins > HCB > chlordane > lindane > mirex [72]. In the soils of Rwanda, for example [73], the occurrence of different organochlorine pesticide residues (DDT) has been reported, with concentrations reaching 120 µg/kg, exceeding the maximum permissible limits. In a case study by Degrendele et al. [74], nine current-use pesticides (CUPs) were identified in agricultural soils in South Africa. The concentrations of chlorpyrifos, carbaryl, and tebuconazole reached up to 63.6, 1.10, and 0.212 ng g⁻¹, respectively, with each surpassing the acceptable daily intake concentrations of 0.1 × 10⁻⁶, 7.5 × 10⁻⁶, and 0.3 × 10⁻⁶ ng g⁻¹, respectively [74]. Another study by Ouhajjou et al. [75], conducted in the Talda Region of Morocco, reported pesticide residues in soils under intensive agriculture with concentrations of up to 348 µg/kg. These pesticides cause the death of essential soil organisms, which play a critical role in nutrient cycling and the maintenance of soil structure. With regard to soil pollution by pesticides, processes such as leaching, surface runoff, sedimentation, and soil filtering capacity control the fate of pesticides in agricultural soils [76]. Most of these pesticide residues end up in groundwater, posing a serious risk to water quality, further exacerbating the challenge of sustainable agriculture [77].

2.5. Soil Pollution and Impacts from Petroleum Hydrocarbons in Africa

Crude oil is a natural resource needed for the economic development of many African countries. However, it also poses a significant risk of soil contamination, with numerous cases having been reported [78,79]. Spillage of crude oil is a major concern with regard to soil health because of its complex composition, containing a mixture of different types of aliphatic and aromatic hydrocarbons [80]. A comparative study of polycyclic aromatic hydrocarbons (PAHs) in soils from Nigeria and Ghana indicates that research in Ghana is limited compared to Nigeria [78]. The latter also reveals that Nigerian soils, especially in the Niger Delta region, have some of the highest concentrations of PAHs. For example, Emoyan et al. [81] analyzed 160 soil samples from ten different sites in Sapele city (in the Niger Delta region) and found total PAH concentrations ranging from 60.76 to 271.11 mg/kg. Additionally, a study by Muze et al. [82] revealed total petroleum hydrocarbon (TPH) concentrations varying from less than 1.00 to 38.32 mg/kg. An extensive review has been performed on soil contamination by petroleum hydrocarbons in various regions across Nigeria [78].

2.6. Soil Contamination by Radionuclides

Radiation emitted by radionuclides poses a significant health risk to humans, potentially causing serious conditions, such as skin burns, cardiovascular diseases, and increased cancer risk. In Africa, radionuclide contamination of soil arises from both natural sources and anthropogenic activities, such as mining operations. Over the past decades, there has been a growing interest in the mining of radionuclides in Africa, especially uranium [83,84]. According to Dasnois [83], Namibia, Niger, Malawi, South Africa, Tanzania, and Botswana are the major African countries that have been actively mining uranium. In general, studies on the assessment of soil contamination by radionuclides have paid much attention to 226Ra, 232Th, 40K, and 238U/235U, compared to other radionuclides, such as 137Cs, 60Co, 214Bi, 208Tl, and gamma radiation. With regard to natural sources, the concentrations of these radionuclides or the activity of the emitted radiation greatly depends on the composition and mineralogy of the underlying parent rock or ore [84,85,86]. For example, Ntsohi et al. [84] found that the isotopic and activity concentrations of 232Th, 238U, 235U, and 232Th/238U in soil samples from a mining site in Botswana varied as a function of the mineralogy of the uranium ore. In the volcanic landscape of the Canary Islands in the NW African Coast, Arnedo et al. [85] analyzed 350 surface (0–5 cm) soil samples and found that the average activity concentrations of 226Ra, 232Th, and 40K were 25.2 Bq/kg, 28.9 Bq/kg, and 384.4 Bq/kg, respectively. However, activity concentrations varied significantly in soil samples depending on the composition of the parent material, with some concentrations reaching very high values of 1214 Bq/kg in areas dominated by calciocarbonatite rocks [85]. In another study conducted on highly weathered soils rich in Nb, Th, and rare earth elements in the Mrima Hills on Kenya’s south coast, a study [86] found concentrations of 232Th, which were >8 times higher than the global average value for soil, indicating that the soils were highly polluted. Similarly, in the Ekiti state of Southwest Nigeria, a study [87] evaluated the activity concentrations of Radon gas from soil and water samples and found that some samples had doses higher than the recommended limit, with the potential to pollute groundwater. With regard to the contribution of anthropogenic sources of radioactive elements, a study assessed the activity concentrations of 238U, 232Th, and 40K in soils around a gold mine tailing in Gauteng province, South Africa, and found that soils with the highest concentrations of radioactive elements were those located in close proximity to the tailings [88]. Similar results have been reported by another study, where anthropogenic activities contribute to the radionuclide contamination of soils [89].

3. Uses of Soil Pollution Maps in Soil Health Monitoring

Soil pollution maps are a highly effective tool for remediation, as they visually represent the spatial distribution of contaminants, highlight areas of high concentration and hotspots, and provide easily interpretable data for decision makers. They facilitate digital sharing among project teams and support coordinated management efforts. In the African context, where soil contamination is often widespread but poorly documented, these maps are particularly valuable for identifying polluted areas, assessing risks to communities, especially those near industrial or waste disposal sites, and guiding targeted interventions to protect human health and prioritize remediation efforts. Some specific insights on the applicability of soil pollution maps for soil health monitoring in selected African countries are given in the Table 1.

Table 1. Case studies on the applicability of soil pollution maps for soil health monitoring.

Country	Soil Pollutants Mapped	Application of Pollution Maps for Soil Health Monitoring	Reference
Nigeria	Petroleum products	The Nigerian government, in partnership with the World Bank, has developed a soil pollution map to identify areas contaminated with petroleum products. This map informs remediation efforts and policy decisions including provision of real-time data on contaminated areas.	[90]
Morocco	Excess salts	An inventory of studies conducted over a quarter century was made to assess drivers of salinity loading in Tadla (Morocco), including irrigation water. An in-depth analysis of the salinity maps generated over many years permitted the authors to identify future areas of investigation for an effective monitoring and management of soil salinity in the area.	[91]
Ghana	Heavy metals	Researchers at the University of Ghana have used satellite-based soil mapping to identify areas contaminated with heavy metals. They used block chain technology to track and verify soil pollution data from electronic waste (e-waste) recycling activities. This information informs policy decisions and remediation efforts.	[92]
Ghana	Organic contaminants and heavy metals	Researchers analyzed various pollutants in surface soil samples obtained from the e-waste processing and dumping sites in Accra (Ghana) to identify pollutants related to e-waste handling.	[93]
Kenya	Pesticides and fertilizers	Crowdsourced soil monitoring using mobile applications and community involvement in Lake Victoria Basin, Kenya. Farmers use low-cost sensors and mobile apps to test for pesticides, fertilizers, and industrial waste contamination in soils. This program includes soil testing and mapping to identify areas with nutrient deficiencies.	[94]
Cameroon	Heavy metals	Researchers attempted to assess the spatial distribution of heavy metals in an industrial zone (Douala) in order to establish pollution indices that could be useful in guiding government action with regard to remediation.	[95]
North Africa	Various elements and pollutants	The Environment Agency—Abu Dhabi (EAD) soil quality monitoring program. The program, launched in 2018, systematically monitors soil quality across the Middle East and North Africa, comprising 664 sites and analyses over 1376 soil samples to screen more than 35 elements and pollutants. The initiative makes use of GIS techniques and statistical analyses to optimize sampling design.	[96]
Africa	Lead and Mercury	The Pure earth project, the first of its kind in developing evidence-based solutions to Hg and Pb contamination, includes a site location mapping feature that is accessible through an interactive map on its website. Once the sites are identified, the most effective methods for reversing trace element contamination are then evaluated. The initiative has implemented remediation processes across 29 trial sites in 17 African countries, including Burkina Faso, Cameroon, Côte d’Ivoire, and numerous other areas in SSA.	[97]
Tanzania	Heavy metals	Researchers established heavy metal pollution maps in the central Dodoma Region of Tanzania using kriging interpolation techniques. The pollution maps permitted the identification of potential sampling points for effective monitoring purposes and enhanced environmental management practices in the region.	[98]
Namibia	Heavy metals	Researchers assessed the dispersion of dust and SO2 emissions from the Tsumeb smelter area, Oshikoto Region, to map contamination and health risk. Pollution maps generated by contouring and kriging techniques are recommended for use by different stakeholders in the sustainable management of ecosystems.	[99]

Table 1. Case studies on the applicability of soil pollution maps for soil health monitoring.

Country	Soil Pollutants Mapped	Application of Pollution Maps for Soil Health Monitoring	Reference
Nigeria	Petroleum products	The Nigerian government, in partnership with the World Bank, has developed a soil pollution map to identify areas contaminated with petroleum products. This map informs remediation efforts and policy decisions including provision of real-time data on contaminated areas.	[90]
Morocco	Excess salts	An inventory of studies conducted over a quarter century was made to assess drivers of salinity loading in Tadla (Morocco), including irrigation water. An in-depth analysis of the salinity maps generated over many years permitted the authors to identify future areas of investigation for an effective monitoring and management of soil salinity in the area.	[91]
Ghana	Heavy metals	Researchers at the University of Ghana have used satellite-based soil mapping to identify areas contaminated with heavy metals. They used block chain technology to track and verify soil pollution data from electronic waste (e-waste) recycling activities. This information informs policy decisions and remediation efforts.	[92]
Ghana	Organic contaminants and heavy metals	Researchers analyzed various pollutants in surface soil samples obtained from the e-waste processing and dumping sites in Accra (Ghana) to identify pollutants related to e-waste handling.	[93]
Kenya	Pesticides and fertilizers	Crowdsourced soil monitoring using mobile applications and community involvement in Lake Victoria Basin, Kenya. Farmers use low-cost sensors and mobile apps to test for pesticides, fertilizers, and industrial waste contamination in soils. This program includes soil testing and mapping to identify areas with nutrient deficiencies.	[94]
Cameroon	Heavy metals	Researchers attempted to assess the spatial distribution of heavy metals in an industrial zone (Douala) in order to establish pollution indices that could be useful in guiding government action with regard to remediation.	[95]
North Africa	Various elements and pollutants	The Environment Agency—Abu Dhabi (EAD) soil quality monitoring program. The program, launched in 2018, systematically monitors soil quality across the Middle East and North Africa, comprising 664 sites and analyses over 1376 soil samples to screen more than 35 elements and pollutants. The initiative makes use of GIS techniques and statistical analyses to optimize sampling design.	[96]
Africa	Lead and Mercury	The Pure earth project, the first of its kind in developing evidence-based solutions to Hg and Pb contamination, includes a site location mapping feature that is accessible through an interactive map on its website. Once the sites are identified, the most effective methods for reversing trace element contamination are then evaluated. The initiative has implemented remediation processes across 29 trial sites in 17 African countries, including Burkina Faso, Cameroon, Côte d’Ivoire, and numerous other areas in SSA.	[97]
Tanzania	Heavy metals	Researchers established heavy metal pollution maps in the central Dodoma Region of Tanzania using kriging interpolation techniques. The pollution maps permitted the identification of potential sampling points for effective monitoring purposes and enhanced environmental management practices in the region.	[98]
Namibia	Heavy metals	Researchers assessed the dispersion of dust and SO2 emissions from the Tsumeb smelter area, Oshikoto Region, to map contamination and health risk. Pollution maps generated by contouring and kriging techniques are recommended for use by different stakeholders in the sustainable management of ecosystems.	[99]

4. Soil Pollution Mapping Across Africa

4.1. Methods of Soil Pollution Mapping

Pollution mapping starts with soil characterization and sampling in the field, analyses of samples in the laboratory, data analysis and interpretation, and mapping using an appropriate methodology. In the field, in situ measurement of elemental concentrations of metals and metalloids is achieved with the aid of appliances such as portable X-ray fluorescence (XRF). Other field appliances that are commonly used include photoionization detectors which enable in situ measurement of volatile organic compounds, portable spectroradiometers for collecting spectral data for soil and pollutant analysis, handheld XRF, and remote satellite-based VIS [100,101]. Soil pollutants are generally mapped at different scales, including local, regional, national, and continental [102].

Two common methods used for mapping soil pollution include simplistic spatial analysis and geostatistical interpolation methods. These methods are briefly described below.

4.1.1. Manual Delineation

This method is important for simplistic assessment, which simply identifies sampling points that exceed threshold values and maps areas affected by soil pollution [103]. It is also useful for zonal evaluations, which assess sampling results for different zones of the area to be sampled. The area to be sampled is partitioned into polygons, with one polygon for each sampling point. These polygons are made of bisecting lines at the center of adjacent sampling locations.

4.1.2. Geostatistical Interpolation Methods

These methods are based on calculating unbiased estimates at unsampled locations. A challenge with this approach is that transition classes between polluted and unpolluted soil areas are difficult to define [104,105]. They are very costly because the starting point is a traditional grid pattern soil sampling followed by soil chemical analysis in the laboratory and subsequent statistical treatment of the data through the application of geostatistical techniques. The interpolation methods most used are inverse distance weighting (IDW), kriging, and spline [106,107,108,109].

The IDW method of interpolation gives spatially close points more weights than distant points, based on the inverse of distance to power. The method has been effectively used to map the spatial distribution of lead in south Africa [110]. Kriging interpolation is based on a model of stochastic spatial variation, which uses a semi-variogram to model the variability and spatial structure of the data.

Kriging interpolation has been effectively used to generate spatial distribution maps of heavy metals in Tanzania [98].

Spatial analysis and fuzzy classification or fuzzy logic techniques have also been used to estimate the spatial distributions of heavy metals in soil [104]. In addition to the above methods, others include machine learning (ML) approaches. Machine learning algorithms, such as support vector machines (SVMs), multi-layer perceptrons (MLPs), random forests (RFs), and extreme random forests (ExtraTrees), provide advanced computational approaches for modeling complex, non-linear relationships between soil properties and pollutant distributions. These methods can predict contaminant levels across unsampled areas, reduce the need for extensive field sampling, and improve the accuracy and efficiency of soil pollution mapping, especially in regions with sparse or heterogeneous data [105].

Turning bands co-simulation algorithms are also used for multivariate mapping of spatial contamination of heavy metals. The spatial uncertainty of four cross-correlated heavy metals (Mn, Fe, Co, and Pb) has been successfully measured in Botswana using the turning bands co-simulation model [111].

4.2. Case Studies on Soil Pollution Mapping Across Africa—Methodology

In conducting this review, relevant publications from the Web of Science, Scopus, and Google Scholar databases were identified using specific search terms, including, soil AND pollution AND mapping, AND “heavy metals OR trace elements OR pesticides OR salinity OR organic pollutants OR hydrocarbons OR radionuclides OR radioactivity OR microplastics”; soil AND pollution AND spatial distribution AND “heavy metals OR trace elements OR pesticides OR salinity OR organic pollutants OR hydrocarbons OR radionuclides OR radioactivity OR microplastics”; soil AND pollution AND maps AND “heavy metals OR trace elements OR pesticides OR salinity OR organic pollutants OR hydrocarbons OR radionuclides OR radioactivity OR microplastics”. In addition to this search procedure, other non-indexed sources, such as university websites, were manually searched using the Google search tool. The search included exclusively scientific articles that have been conducted within African countries and published between 2005 and 2024, in both English and French. This period was selected based on the advent and rise of DSM. In selecting the papers, the Mendeley Reference Manager (version 2.136.0) was used and included the following process: identification, screening, eligibility, and inclusion (Figure 1). The identification stage mainly involved searching for the key terms as described above from the various databases. Screening was performed by selecting articles based on the title, the geographical location and the content of the abstract. The eligibility criterion was based on the reading of the entire paper (notably the methodology and results sections) to verify if there was a mapping procedure and the production of a pollution map. Lastly, duplications were excluded and articles that were retained (included) were those that had at least one of the aforementioned pollutants, a mapping methodology, a pollution map, and conducted in an area within Africa.

The search resulted in a total of 4066 articles for “heavy metals” + “trace metals” + “trace elements”, 116 articles for “pesticides”, 08 articles for “radionuclides”, 1923 articles for “hydrocarbons” + “persistent organic pollutants”, 53 articles for “salinity”, and 01 article for “microplastics”. After the screening process, a total of 289 potential articles were recorded. After reading the methodology and result sections of each paper to identify those that specifically included a pollution map, a total of 70 papers were found eligible and used for the review. Detailed information about the papers (pollution type, location, country, scale, mapping method, validation method and validation statistics) are shown in Table 2.

The table summarizes soil pollution mapping efforts across Africa, highlighting a wide range of pollutants, including heavy metals, pesticides, hydrocarbons, radionuclides, and excess salts. Studies vary in scale from local to continental, with sampling sizes ranging from single digits to tens of thousands of observations. Dominant mapping methods include digital soil mapping (DSM), hybrid approaches, geoelectrical mapping, and compositional mapping (CM), often accompanied by validation techniques such as cross-validation, data splitting, or out-of-bag testing, where reported. While many studies focus on heavy metals at local scales in countries such as Nigeria, Egypt, Ghana, and Morocco, others address national or regional assessments of excess salts, radionuclides, and persistent organic pollutants. Accuracy metrics, where available, indicate generally high performance for DSM and hybrid methods, though gaps exist due to missing data or unreported validation. Overall, the dataset reveals substantial spatial coverage and methodological diversity but also highlights uneven geographic representation in countries such as Burkina Faso, Congo, Namibia, and most of Central Africa, limited reporting of model performance, and underexplored pollutants in African soil mapping studies.

In general, most soil pollution maps established in different areas across Africa have been created on a local scale (80%), outnumbering those established at regional (13%), national (6%), or continental scales (1%). Detailed surveys are usually very demanding in various aspects including soil sampling and laboratory analyses. The high financial costs involved in conducting extensive field surveys and analyzing many soil samples are one of the major reasons for limiting the studies at the local scale.

Table 2. Examples of soil pollution maps, scales, and mapping methods across Africa.

Pollutant	Scale	No. of Samples	Location	Country	Mapping Method	Accuracy Method	R2	RMSE	Reference
Free cyanide	Local	73	Zougnazagmiline (northern part of Burkina Faso) and Galgouli (southern part of Burkina Faso)	Burkina Faso	DSM	n.a.	n.a.	n.a.	[112]
Heavy metals	Local	103	Central Dodoma Region	Tanzania	DSM	n.a.	n.a.	n.a.	[98]
Heavy metals	Local	552	Nangodi area, Talensi-Nabdam district, Upper East Region	Ghana	DSM	Generalized Cross-V	0.98–0.99	0.12–1.04	[113]
Heavy metals	Local	86	Sohag Governorate	Egypt	DSM	n.a.	n.a.	n.a.	[114]
Heavy metals	Local	101	Annaba	Algeria	DSM	Information theoretic approach	0.45–0.64	-	[115]
Heavy metals	Local	60	El-Minia Governorate	Egypt	DSM	Cross-V	-	0.50–0.91	[116]
Heavy metals	Local	36	Setif city	Algeria	DSM	n.a.	n.a.	n.a.	[117]
Heavy metals	Local	24	Ogere	Nigeria	DSM	n.a.	n.a.	n.a.	[118]
Heavy metals	Local	33	El-Gharbia Governorate	Egypt	DSM	n.a.	n.a.	n.a.	[119]
Heavy metals	Local	54	Mayanga, Southwest Brazaville	Congo	DSM	n.a.	n.a.	n.a.	[120]
Heavy metals	Local	60	Nairobi	Kenya	DSM	Variogram analysis	n.a.	n.a.	[121]
Heavy metals	Local	198	Usangu Basin	Tanzania	Hybrid	n.a.	n.a.	n.a.	[122]
Heavy metals	Local	40	Abuakwa South Municipal area	Ghana	Hybrid	n.a.	n.a.	n.a.	[123]
Heavy metals	Local	31	Bekao, Mbéré division, Adamawa Region	Cameroon	DSM	Cross-V	0.95	0.43	[124]
Heavy metals	Local	21	Benin city, Edo State	Nigeria	DSM	n.a.	n.a.	n.a.	[125]
Heavy metals	Local	72	Lagos metropolis	Nigeria	DSM	n.a.	n.a.	n.a.	[126]
Heavy metals	Local	60	Yaounde	Cameroon	DSM	n.a.	n.a.	n.a.	[127]
Heavy metals	Local	24	Lala-Manjo highway, Littoral Region	Cameroon	Hybrid	-	-	-	[128]
Heavy metals	Local	28	Bétaré-Oya, East Region	Cameroon	DSM	Cross-V	n.a.	0.96–1.11	[129]
Heavy metals	Local	25	Beni-Moussa, Tadla Plain	Morocco	DSM	n.a.	n.a.	n.a.	[130]
Heavy metals	Local	98	Touiref District	Tunisia	Hybrid	n.a.	n.a.	n.a.	[131]
Heavy metals	Local	109	Fedj Lahdoum	Tunisia	DSM	Cross-V	n.a.	n.a.	[132]
Heavy metals	Local	65	Tsumeb smelter area, Oshikoto Region	Namibia	Hybrid	n.a.	n.a.	n.a.	[99]
Heavy metals	Local	340	Santiago Island	Cape Verde	DSM	Cross-V	n.a.	0.02–5.03	[133]
Heavy metals	Local	16	Ijero-Ekiti	Nigeria	DSM	n.a.	n.a.	n.a.	[134]
Heavy metals	Local	22	Niger Delta basin	Nigeria	NA	n.a.	n.a.	n.a.	[135]
Heavy metals	Local	9	Al-Qalyubia Governorate	Egypt	DSM	n.a.	n.a.	n.a.	[136]
Heavy metals	Local	26	West Girga, Sohag governorate	Egypt	Hybrid	n.a.	n.a.	n.a.	[137]
Heavy metals	Local	24	Northern part of the Nile Delta	Egypt	Hybrid	n.a.	n.a.	n.a.	[138]
Heavy metals	Local	107	Kumasi metropolis	Ghana	n.a.	n.a.	n.a.	n.a.	[139]
Heavy metals	Local	1050	Maibele Airstrip North	Botswana	DSM	Cross-V	n.a.	n.a.	[111]
Heavy metals and nitrate	Local	16	University of Ibadan campus, Ibadan metropolis	Nigeria	Geoelectrical mapping	n.a.	n.a.	n.a.	[140]
Heavy metals and radionuclide (238U, 226Ra, 232Th, 40K, and 210Pb)	Local	20	Northeastern Nile Valley	Egypt	CM	n.a.	n.a.	n.a.	[141]
Heavy metals	Local	n.a.	Egbema Kingdom, Delta State	Nigeria	DSM	n.a.	n.a.	n.a.	[142]
Toxic metals	Local	62	Kettara Mine, west of Marrakech city	Morocco	DSM	Semi-variogram analysis	-	0.79–0.99	[143]
Toxic metals	Local	75	Ikirun	Nigeria	DSM	n.a.	n.a.	n.a.	[144]
Heavy metals	National	60	Confluence of the Nairobi and Thiririka rivers	Kenya	Hybrid	“out of bag” (OOB) testing and Cross-V	0.78–0.83	0.27–0.51	[145]
Heavy metals	National	942	South Africa	South Africa	DSM	n.a.	n.a.	n.a.	[110]
Heavy metals	Regional	159	Nile Valley in Minia Governorate	Egypt	DSM	n.a.	n.a.	n.a.	[146]
Trace metals	Regional	60	Eastern Nile Delta	Egypt	CM	n.a.	n.a.	n.a.	[147]
Hydrocarbons	Local	290	Alode, Eleme Local Government Area of Rivers State	Nigeria	Geoelectrical mapping	Data splitting	n.a.	n.a.	[148]
Metal/metalloid contaminants	Local	196	Nkana copper smelter, Copperbelt Province	Zambia	CM	n.a.	n.a.	n.a.	[149]
Oil spill incidence	National	n.a.	Nigeria	Nigeria	CM	-	-	-	[150]
Persistent organic pollutants	Local	108	Nyabarongo lower catchment	Rwanda	CM				[73]
Pesticides	Local	27	Béré watershed	Ivory Coast	DSM	Cross-V	-	5.16	[151]
Pesticides	Regional	30	Akwa Ibom State	Nigeria	CM	n.a.	n.a.	n.a.	[152]
Pesticides	Regional	32	Southeastern, eastern, southwestern, and central-western regions	Tanzania	Hybrid	n.a.	n.a.	n.a.	[153]
Polycyclic aromatic hydrocarbons	Local	16	Eleme and Ahoada East, Niger Delta	Nigeria	CM	n.a.	n.a.	n.a.	[154]
Polycyclic aromatic hydrocarbons (PAHs)	Local	129	Kumasi metropolis	Ghana	Hybrid	n.a.	n.a.	n.a.	[155]
Radioactive minerals	Local	n.a.	Mrima Hill, South Coast of Kenya	Kenya	DSM	-	-	-	[86]
Radionuclides including Gamma radiation	Local	811	Jos Plateau	Nigeria	DSM	n.a.	n.a.	n.a.	[156]
Radionuclides including 238U, 226Ra, 232Th, and 40K	Local	44	Itu local government area, Akwa Ibom State	Nigeria	Hybrid	-	-	-	[157]
Radionuclides including 238U, 232Th and 40K	Local	n.a.	Rustenburg	South Africa	DSM	n.a.	n.a.	n.a.	[89]
Radionuclides including 238U, 232Th and 40K	Local	163	Witwatersrand Basin, Gauteng Province	South Africa	n.a.	-	-	-	[89]
Radionuclides including 222Rn	Local	100	Ekiti State	Nigeria	DSM	n.a.	n.a.	n.a.	[87]
Radionuclides including 137Cs 60Co, 40K, 214Bi and 208Tl	Regional	n.a.	West Coast (Saldanha Bay nature reserve) and East Coast (near Amanzimtot)	South Africa	Hybrid	n.a.	n.a.	n.a.	[158]
Radionuclides including 226Ra, 232Th and 40K	Regional	350	Canary Islands	Off the Western Sahara African coast	DSM	n.a.	n.a.	n.a.	[85]
Excess salts	Local	25	Tafilalet plain, Errachidia Region	Morocco	Hybrid	Cross-V	0.93	n.a.	[159]
Excess salts	Local	51	Fatnassa Oasis	Tunisia	Hybrid	Data splitting	0.56	2.94	[160]
Excess salts	Local	229	Zaghouan Governorate	Tunisia	DSM	Data splitting	0.67	0.12	[161]
Excess salts	Local	45	Zelfana municipality, Ghardaïa province	Algeria	DSM	Data splitting	n.a.	1.94–7.16; 1.95–7.45	[162]
Excess salts	Local	685	Beni Amir, Tadla plain	Morocco	Hybrid	n.a.	n.a.	n.a.	[163]
Excess salts	Local	20	El-Salhia Area, East of Nile Delta	Egypt	DSM	n.a.	n.a.	n.a.	[164]
Excess salts	Local	420	Kollo, Southeast of Niamey	Niger	DSM			0.79–0.83	[165]
Excess salts	Regional	n.a.	Tadla	Morocco	DSM	Cross-V	0.55–0.97	0.08–2.35	[91]
Excess salts	Regional	92	Beni Amir, Tadla plain	Morocco	DSM	Cross-V	n.a.	0.42	[166]
Excess salts	Regional	92	East of the Nile Delta	Egypt	DSM	Cross-V	-	0.38–0.39	[167]
Excess salts	National	1083	Cameroon	Cameroon	DSM	Data splitting	n.a.	n.a.	[168]
Excess salts	Continental	>60,000	Africa	Africa	DSM	Data splitting	0.74–0.76	0.56–1.04	[21]
Total N and P loads	Local	n.a.	Muanza city	Tanzania	DSM	n.a.	n.a.	n.a.	[169]

Cross-V = cross-validation; CM = conventional mapping (including contouring, GPS survey, field survey); DSM = digital soil mapping (including geostatistical and ML approaches); hybrid = a combination of two or more mapping techniques including field and GPS survey, remote sensing, and DSM, contouring and DSM, multi-height electromagnetic induction and quasi-3d inversion algorithms; n.a. = not available.

Table 2. Examples of soil pollution maps, scales, and mapping methods across Africa.

Pollutant	Scale	No. of Samples	Location	Country	Mapping Method	Accuracy Method	R2	RMSE	Reference
Free cyanide	Local	73	Zougnazagmiline (northern part of Burkina Faso) and Galgouli (southern part of Burkina Faso)	Burkina Faso	DSM	n.a.	n.a.	n.a.	[112]
Heavy metals	Local	103	Central Dodoma Region	Tanzania	DSM	n.a.	n.a.	n.a.	[98]
Heavy metals	Local	552	Nangodi area, Talensi-Nabdam district, Upper East Region	Ghana	DSM	Generalized Cross-V	0.98–0.99	0.12–1.04	[113]
Heavy metals	Local	86	Sohag Governorate	Egypt	DSM	n.a.	n.a.	n.a.	[114]
Heavy metals	Local	101	Annaba	Algeria	DSM	Information theoretic approach	0.45–0.64	-	[115]
Heavy metals	Local	60	El-Minia Governorate	Egypt	DSM	Cross-V	-	0.50–0.91	[116]
Heavy metals	Local	36	Setif city	Algeria	DSM	n.a.	n.a.	n.a.	[117]
Heavy metals	Local	24	Ogere	Nigeria	DSM	n.a.	n.a.	n.a.	[118]
Heavy metals	Local	33	El-Gharbia Governorate	Egypt	DSM	n.a.	n.a.	n.a.	[119]
Heavy metals	Local	54	Mayanga, Southwest Brazaville	Congo	DSM	n.a.	n.a.	n.a.	[120]
Heavy metals	Local	60	Nairobi	Kenya	DSM	Variogram analysis	n.a.	n.a.	[121]
Heavy metals	Local	198	Usangu Basin	Tanzania	Hybrid	n.a.	n.a.	n.a.	[122]
Heavy metals	Local	40	Abuakwa South Municipal area	Ghana	Hybrid	n.a.	n.a.	n.a.	[123]
Heavy metals	Local	31	Bekao, Mbéré division, Adamawa Region	Cameroon	DSM	Cross-V	0.95	0.43	[124]
Heavy metals	Local	21	Benin city, Edo State	Nigeria	DSM	n.a.	n.a.	n.a.	[125]
Heavy metals	Local	72	Lagos metropolis	Nigeria	DSM	n.a.	n.a.	n.a.	[126]
Heavy metals	Local	60	Yaounde	Cameroon	DSM	n.a.	n.a.	n.a.	[127]
Heavy metals	Local	24	Lala-Manjo highway, Littoral Region	Cameroon	Hybrid	-	-	-	[128]
Heavy metals	Local	28	Bétaré-Oya, East Region	Cameroon	DSM	Cross-V	n.a.	0.96–1.11	[129]
Heavy metals	Local	25	Beni-Moussa, Tadla Plain	Morocco	DSM	n.a.	n.a.	n.a.	[130]
Heavy metals	Local	98	Touiref District	Tunisia	Hybrid	n.a.	n.a.	n.a.	[131]
Heavy metals	Local	109	Fedj Lahdoum	Tunisia	DSM	Cross-V	n.a.	n.a.	[132]
Heavy metals	Local	65	Tsumeb smelter area, Oshikoto Region	Namibia	Hybrid	n.a.	n.a.	n.a.	[99]
Heavy metals	Local	340	Santiago Island	Cape Verde	DSM	Cross-V	n.a.	0.02–5.03	[133]
Heavy metals	Local	16	Ijero-Ekiti	Nigeria	DSM	n.a.	n.a.	n.a.	[134]
Heavy metals	Local	22	Niger Delta basin	Nigeria	NA	n.a.	n.a.	n.a.	[135]
Heavy metals	Local	9	Al-Qalyubia Governorate	Egypt	DSM	n.a.	n.a.	n.a.	[136]
Heavy metals	Local	26	West Girga, Sohag governorate	Egypt	Hybrid	n.a.	n.a.	n.a.	[137]
Heavy metals	Local	24	Northern part of the Nile Delta	Egypt	Hybrid	n.a.	n.a.	n.a.	[138]
Heavy metals	Local	107	Kumasi metropolis	Ghana	n.a.	n.a.	n.a.	n.a.	[139]
Heavy metals	Local	1050	Maibele Airstrip North	Botswana	DSM	Cross-V	n.a.	n.a.	[111]
Heavy metals and nitrate	Local	16	University of Ibadan campus, Ibadan metropolis	Nigeria	Geoelectrical mapping	n.a.	n.a.	n.a.	[140]
Heavy metals and radionuclide (238U, 226Ra, 232Th, 40K, and 210Pb)	Local	20	Northeastern Nile Valley	Egypt	CM	n.a.	n.a.	n.a.	[141]
Heavy metals	Local	n.a.	Egbema Kingdom, Delta State	Nigeria	DSM	n.a.	n.a.	n.a.	[142]
Toxic metals	Local	62	Kettara Mine, west of Marrakech city	Morocco	DSM	Semi-variogram analysis	-	0.79–0.99	[143]
Toxic metals	Local	75	Ikirun	Nigeria	DSM	n.a.	n.a.	n.a.	[144]
Heavy metals	National	60	Confluence of the Nairobi and Thiririka rivers	Kenya	Hybrid	“out of bag” (OOB) testing and Cross-V	0.78–0.83	0.27–0.51	[145]
Heavy metals	National	942	South Africa	South Africa	DSM	n.a.	n.a.	n.a.	[110]
Heavy metals	Regional	159	Nile Valley in Minia Governorate	Egypt	DSM	n.a.	n.a.	n.a.	[146]
Trace metals	Regional	60	Eastern Nile Delta	Egypt	CM	n.a.	n.a.	n.a.	[147]
Hydrocarbons	Local	290	Alode, Eleme Local Government Area of Rivers State	Nigeria	Geoelectrical mapping	Data splitting	n.a.	n.a.	[148]
Metal/metalloid contaminants	Local	196	Nkana copper smelter, Copperbelt Province	Zambia	CM	n.a.	n.a.	n.a.	[149]
Oil spill incidence	National	n.a.	Nigeria	Nigeria	CM	-	-	-	[150]
Persistent organic pollutants	Local	108	Nyabarongo lower catchment	Rwanda	CM				[73]
Pesticides	Local	27	Béré watershed	Ivory Coast	DSM	Cross-V	-	5.16	[151]
Pesticides	Regional	30	Akwa Ibom State	Nigeria	CM	n.a.	n.a.	n.a.	[152]
Pesticides	Regional	32	Southeastern, eastern, southwestern, and central-western regions	Tanzania	Hybrid	n.a.	n.a.	n.a.	[153]
Polycyclic aromatic hydrocarbons	Local	16	Eleme and Ahoada East, Niger Delta	Nigeria	CM	n.a.	n.a.	n.a.	[154]
Polycyclic aromatic hydrocarbons (PAHs)	Local	129	Kumasi metropolis	Ghana	Hybrid	n.a.	n.a.	n.a.	[155]
Radioactive minerals	Local	n.a.	Mrima Hill, South Coast of Kenya	Kenya	DSM	-	-	-	[86]
Radionuclides including Gamma radiation	Local	811	Jos Plateau	Nigeria	DSM	n.a.	n.a.	n.a.	[156]
Radionuclides including 238U, 226Ra, 232Th, and 40K	Local	44	Itu local government area, Akwa Ibom State	Nigeria	Hybrid	-	-	-	[157]
Radionuclides including 238U, 232Th and 40K	Local	n.a.	Rustenburg	South Africa	DSM	n.a.	n.a.	n.a.	[89]
Radionuclides including 238U, 232Th and 40K	Local	163	Witwatersrand Basin, Gauteng Province	South Africa	n.a.	-	-	-	[89]
Radionuclides including 222Rn	Local	100	Ekiti State	Nigeria	DSM	n.a.	n.a.	n.a.	[87]
Radionuclides including 137Cs 60Co, 40K, 214Bi and 208Tl	Regional	n.a.	West Coast (Saldanha Bay nature reserve) and East Coast (near Amanzimtot)	South Africa	Hybrid	n.a.	n.a.	n.a.	[158]
Radionuclides including 226Ra, 232Th and 40K	Regional	350	Canary Islands	Off the Western Sahara African coast	DSM	n.a.	n.a.	n.a.	[85]
Excess salts	Local	25	Tafilalet plain, Errachidia Region	Morocco	Hybrid	Cross-V	0.93	n.a.	[159]
Excess salts	Local	51	Fatnassa Oasis	Tunisia	Hybrid	Data splitting	0.56	2.94	[160]
Excess salts	Local	229	Zaghouan Governorate	Tunisia	DSM	Data splitting	0.67	0.12	[161]
Excess salts	Local	45	Zelfana municipality, Ghardaïa province	Algeria	DSM	Data splitting	n.a.	1.94–7.16; 1.95–7.45	[162]
Excess salts	Local	685	Beni Amir, Tadla plain	Morocco	Hybrid	n.a.	n.a.	n.a.	[163]
Excess salts	Local	20	El-Salhia Area, East of Nile Delta	Egypt	DSM	n.a.	n.a.	n.a.	[164]
Excess salts	Local	420	Kollo, Southeast of Niamey	Niger	DSM			0.79–0.83	[165]
Excess salts	Regional	n.a.	Tadla	Morocco	DSM	Cross-V	0.55–0.97	0.08–2.35	[91]
Excess salts	Regional	92	Beni Amir, Tadla plain	Morocco	DSM	Cross-V	n.a.	0.42	[166]
Excess salts	Regional	92	East of the Nile Delta	Egypt	DSM	Cross-V	-	0.38–0.39	[167]
Excess salts	National	1083	Cameroon	Cameroon	DSM	Data splitting	n.a.	n.a.	[168]
Excess salts	Continental	>60,000	Africa	Africa	DSM	Data splitting	0.74–0.76	0.56–1.04	[21]
Total N and P loads	Local	n.a.	Muanza city	Tanzania	DSM	n.a.	n.a.	n.a.	[169]

Cross-V = cross-validation; CM = conventional mapping (including contouring, GPS survey, field survey); DSM = digital soil mapping (including geostatistical and ML approaches); hybrid = a combination of two or more mapping techniques including field and GPS survey, remote sensing, and DSM, contouring and DSM, multi-height electromagnetic induction and quasi-3d inversion algorithms; n.a. = not available.

The number of soil samples used in establishing soil maps is highly variable throughout the various studies. The majority of the studies make use of <100 or between 100 and 900 soil samples, alongside varying sampling areas. This leads to a high variability in sampling density, but which is generally very low [170]. Very few studies used sample sizes > 1000 samples. The largest dataset used included about 60,000 soil sampling points, consistent with the continental scale at which the study was conducted [21].

Across Africa, soil pollution maps are established using a variety of methods regardless of the scale. The different methods evaluated from the available literature indicate that digital mapping methods such as kriging interpolation techniques are the most used, followed by a combination of methods (hybrid) including DSM and conventional mapping, while geoelectrical methods (notably geoelectrical soil mapping and electrical-resistivity imaging) are the least used (Figure 2). One of the factors favoring the use of these digital techniques is that most of them, especially kriging interpolation, are incorporated as analytical tools in many GIS software tools, which are readily available to users. In addition, ML methods are also commonly used in combination with other methods to produce soil pollution maps.

Regarding the types of soil pollutants, heavy metals are the most frequently evaluated across the continent (57%) (Figure 3), particularly in countries within the humid tropics such as Nigeria, Ghana, and Cameroon. In the humid tropics where acidic and nutrient-poor soils dominate, there is a high usage of mineral fertilizers and other chemical inputs such as pesticides, which are a major source of heavy metals in soils [50]. In addition, the indiscriminate dumping of municipal and industrial waste without prior treatment contributes to heavy metal accumulation in soils. Following heavy metals, excess salts (salinity) (16%) are extensively monitored in arid and semi-arid environments of Northern Africa, where salts naturally accumulate from the weathering of the prevailing saline parent materials. Morocco has recorded one of the highest number of studies involving mapping of soil salinity [91]. Other countries that have generated soil salinity maps include Tunisia, Egypt, Algeria, and Niger (Table 2). Other soil pollutants that have received considerable attention and been mapped include radionuclides (12%), with South Africa having the highest number of studies (Table 2).

Mining is a vital part of South Africa’s economy, and mining activities are known to increase the levels of naturally occurring radioactive materials in the environment. Thus, there is a need to monitor the activity of radionuclides around mines to establish strategies to mitigate soil degradation. According to the data in Table 2, there are very few maps on the distribution of pesticides (4%) and persistent organic pollutants (6%). This could be due to the limited laboratory facilities for the analysis of pesticide residues in soil samples. Where analyses of pesticides are conducted, they are usually performed at the sites where the chemicals are either stored or applied. Again, the spatial distribution of these pollutants, which might be affected by climatic parameters such as rainfall, temperature, and wind, is hardly assessed.

The number of studies conducted per country, based on the data in Table 2, is shown in Figure 4, with Nigeria exhibiting the highest count. This can be explained by the comparatively large population of Nigeria, a very high number of research institutions, and the rapid urbanization of the country, which is accompanied by increased waste production [171]. Thus, there is a need to assess environmental pollution.

4.2.1. Limitations in Mapping Methods—A Critical Analysis

Soil pollution mapping is essential for monitoring environmental health, guiding land management and informing remediation efforts. However, several challenges limit the feasibility, accuracy, and effectiveness of soil pollution mapping methods, especially in developing regions such as Africa. The challenges range from accuracy assessment, data scarcity, and poor infrastructure to limitations in remote sensing and ML approaches. Here, we give a critical analysis of some of these challenges.

4.2.2. Accuracy Assessment

From the studies reported in Table 2 and Figure 2, the most used method for generating soil pollution maps is digital soil mapping techniques. As mentioned in previous studies [170], these methods face many challenges, especially when it comes to their accuracy. In general, most of the studies that used kriging or IDW interpolation did not report any statistics to assess the accuracy of the interpolation methods used. For example, the establishment of a national-scale spatial distribution map of Mehlich-3 extractable lead in surface soils in South Africa used the IDW method without reporting any statistics for accuracy assessment [110], a pattern detected in other studies [119,120,125]. On the other hand, only a few studies have reported accuracy assessment statistics for the interpolated method used. However, even when reported, there is usually great disparity in the number and type of statistics employed. For example, Elvine Paternie et al. [129] established spatial distribution maps of heavy metals in East Cameroon using kriging interpolation methods. In their study, they reported the theoretical semi-variogram model values (e.g., nugget) and the estimation error using RMSE, average standardized error, and standardized means. In a similar study conducted to assess the spatial distribution of heavy metals in soils of Central Dodoma Region in Tanzania, Ref. [98] used the kriging interpolation technique and reported that the Gaussian semi-variogram model was used. However, no statistics were reported to assess model accuracy. In general, when it comes to geostatistical mapping, the reliability of interpolation methods has to meet some basic requirements, such as accuracy, precision, estimation power, ability to handle huge datasets, computational efficacy, and flexibility [172]. In most cases, most of the soil pollution maps generated across Africa rely on very small datasets for relatively large surface areas, such as the studies conducted in the Talda plain in Morocco [130], in Tanzania [153], and in Nigeria [134].

Given the difficulty in identifying a specific method that can meet all the aforementioned requirements, it is imperative that different methods be used within a particular study to evaluate which of them can generate results that are closest to the real situation. Unfortunately, a large majority of the studies rely on a single interpolation method without justifying the choice of the method. Only a few studies have compared the effectiveness of at least two mapping methods [111,124,162]. Other studies compared the efficiency of IDW and kriging methods for soil salinity mapping in southern Algeria by calculating the mean error and RMSE, while considering the mapping scales and sampling density [162]. Based on their analysis, they found that the IDW method was more efficient than kriging for predicting the spatial distribution of soil salinity and recommended the method for use in arid and semi-arid environments characterized by high salt content and stagnant drainage water. In the study, turning bands co-simulation algorithm and kriging techniques were used to map heavy metals in a semiarid Ni–Cu exploration field in Botswana [111]. The two methods were validated using a cross-validation technique, and based on the analysis of method accuracy, the authors recommended the use of turning bands co-simulation methods for mapping heavy metal uncertainty in areas with similar geochemical properties. At an abandoned gold mining site along the Lom River in Bekao, Adamawa Region, Cameroon, a study applied a combination of geostatistics and ML approaches to predict trace metal concentrations in sediments. In their work, 31 surface samples were analyzed to assess the levels of As, Cr, Cu, Fe, Mn, Ni, Pb, Sn, and Zn [124]. The study compared the performance of ordinary kriging (OK), ordinary co-kriging (OCK), and ML (ANN) models. Based on model validation statistics, the ANN model demonstrated superior predictive capabilities, suggesting that ML approaches can enhance the accuracy of spatial predictions. However, the study also highlighted challenges such as data scarcity, which can affect the reliability of ML models, and the need for high-quality training data to improve model performance. From the various accuracy methods and statistics shown in Table 2, it appears that, in most cases, there is a mismatch between sample size/density and mapping method, which can lead to the generation of inaccurate spatial information, with potentially severe consequences vis à vis decision making.

Out of the 70 studies reported in Table 2, only 13% used validation methods to assess accuracy or effectiveness of mapping methods, notably coefficients of determination (R2) and root mean square error (RMSE). For the few studies that used coefficient of determination as a criterion for accuracy evaluation, mean values ranged between 0.45 and 0.99 (mean = 0.76 ± 0.05, CV = 24.2%). Root mean square error (RMSE) values, on the other hand, ranged between 0.02 and 7.45 (mean = 1.83 ± 0.41, CV = 115.2%). A closer look at studies that used R2 values indicated that the best model (R2 = 0.99) was obtained in a local-scale study in Ghana that used 552 data points alongside ML as mapping method for heavy metals [113]. This could indicate that the sampling density and mapping method were appropriate for the local-scale study. On the other hand, a nation-wide study in Kenya used only 60 sampling points alongside ML techniques for mapping heavy metals and obtained R2 values between 0.78 and 0.83, which are considered good. An analysis of these two case studies indicates the complexity of relating sampling density to mapping scale.

4.2.3. Data Quality and Availability

One of the most significant barriers to effective soil pollution mapping in Africa is the lack of reliable and high-resolution data. Soil pollution assessments require comprehensive datasets that include heavy metal concentrations, organic pollutants, and other contaminants. However, in many African countries limited soil monitoring networks exist compared to Europe [173], leading to outdated and unreliable datasets being used for predicting soil properties.

The effectiveness of both geostatistical and ML methods heavily depends on the availability of high-quality data. In Africa, challenges such as low sampling density and spatial clustering of soil data limit the reliability of these techniques. The reliance on legacy soil datasets, which may lack comprehensive spatial coverage, can lead to unreliable predictions and misinformed decision-making [170]. Additionally, inadequate infrastructure hampers data collection, processing and dissemination efforts [170]. Remote sensing and geospatial analysis was used in Agbogbloshie, Ghana, one of the world’s largest e-waste dumping sites with severe soil contamination, high concentrations of lead, cadmium, and other toxic metals while. However, due to inadequate ground-truth data, some pollution hotspots were underestimated or overestimated, illustrating the challenge of validating remote sensing results.

4.2.4. Reliability of Machine Learning Outputs

ML techniques have emerged as powerful tools for soil attributes mapping [174]. ML models, while powerful, often produce outputs that are difficult for practitioners to interpret and apply. The lack of transparency in these models can hinder their practical application in soil pollution mapping, especially when local expertise in data science is limited [175]. Even though their application in Africa is hindered by model interpretability issues, data scarcity, and infrastructure constraints, there are prospects for addressing such challenges, given that recent advances in digital soil mapping allow for efficient interpretability in ML models [176,177]. Additionally, recent advances in digital mapping have introduced novel interpretation methods and ML analyses for soil pollution data [178]. These advancements have demonstrated that ML yields reliable model results and that challenges such as data sparsity are no longer a valid justification for the omission of quantitative impact predictions, especially in soil pollution assessment. A local-scale study in Ghana employed ML to map heavy metals, and based on the prediction accuracy (using the R2 values), the authors could easily identify the best ML models for heavy metals mapping [113]. Similar results were obtained when ML was used for soil salinity mapping at the continental scale [21]. In another study conducted to map rare earth elements in the Mrima Hills on Kenya’s south coast, a study used the maximum likelihood classifier to accurately infer the presence of radioactive elements in soil samples [86]. The authors reported an overall classification accuracy of 91% using soil/rocks spectral signatures and demonstrated the potential of such methods for radioactive mapping, especially in resource-constrained environments. Thus, ML appears to be a versatile tool for the production of reliable soil pollution maps.

4.2.5. Effectiveness of Digital Mapping Methods

Methods such as geostatistics have been widely applied to spatial studies in soil and agronomic research, including soil pollution mapping. Basically, geostatistics provides descriptive tools such as semivariograms to characterize the spatial pattern of continuous and categorical soil attributes such as heavy metal concentrations, salinity levels, and radionuclide activity concentrations. The results depicted in Table 2 and Figure 2 show that kriging and inverse distance weighting (IDW) are the most common interpolation techniques used for predicting attribute values at unsampled locations. With regard to soil pollution mapping, geostatistics plays an important role in assessing the uncertainty associated with unsampled values, often in the form of probability maps indicating the likelihood of exceeding critical threshold values or permissible limits [179]. Unfortunately, most (about 70%) of the studies using geostatistical methods do not assess the prediction accuracy nor the spatial uncertainty of the predicted soil pollution attributes (Table 2). Even when model performance is evaluated, accuracy metrics such as the root mean squared error (RMSE) or the coefficient of determination (R2) are hardly reported. In the absence of such important information (such as probability maps), it is challenging to assess the reliability and effectiveness of the generated soil information products (pollution maps), which in turn limits their practical application in decision-making processes, such as delineating contaminated areas [179]. Two major challenges limiting the effectiveness of geostatistical mapping of soil attributes are data quality and sampling density [180]. The effective implementation of geostatistical methods requires extensive soil sampling and laboratory analyses, leading to significant financial and logistical burdens. In Sub-Saharan Africa, the scarcity of local laboratories capable of analyzing a broad spectrum of contaminants exacerbates this issue. Consequently, samples often need to be sent abroad for analysis, further increasing costs and complicating logistics [4]. A case study in South Africa investigated heavy metal dynamics and pollution in the vicinity of a landfill [181]. The study employed various techniques, including spectroscopic methods, to assess the contamination levels of various metals. The research highlighted the challenges of using advanced analytical methods in regions with limited resources, as these techniques require extensive equipment calibration and laborious laboratory procedures. Additionally, the study emphasized the need for rapid and robust assessment methods that can be easily used for soil pollution monitoring [181]. Therefore, addressing data quality issues and increasing sampling density to support robust geostatistical analysis and uncertainty quantification need to be considered in future studies.

5. Gaps and Challenges in Soil Pollution Mapping with Respect to the Soil Information System Framework in Sub-Saharan Africa

A key barrier to effective soil pollution mapping in Sub-Saharan Africa (SSA) is the fragmented, localized, and inconsistent data that is not integrated into a usable soil information system [52,182], preventing the development of robust regional- or continental-scale models. Data are also rarely collected over time, making it difficult to assess long-term pollution trends and dynamics [183]. Although satellite products such as Landsat 8 and Sentinel-1 could support spatial assessments by providing information on land use and land cover changes [184], their use in soil pollution mapping remains minimal due to limited technical expertise and processing capacity in many SSA countries. Technological and institutional barriers further constrain mapping progress [185], as restricted access to advanced laboratory facilities, computational resources, and skilled personnel hampers efforts to generate detailed pollution datasets [52]. Existing soil inventories typically prioritize agricultural productivity over pollution assessment, leaving critical contaminants such as heavy metals underrepresented. For example, the Africa Soil Information Service (AfSIS) improved soil health mapping but faced difficulties in harmonizing methods and integrating diverse datasets [186].

Another gap lies in the limited research on interactions between multiple pollutants and their cumulative impacts on soil health [182]. Furthermore, the scarcity of integrated approaches that combine scientific data with participatory inputs, including indigenous knowledge [187,188,189,190], reduces the relevance and applicability of mapping outputs for local contexts. Overall, these gaps highlight the urgent need for harmonized methodologies, broader geographic and temporal coverage, and interdisciplinary approaches to strengthen soil pollution mapping in SSA.

5.1. Lack of Political Will and Policy Weaknesses

Weak governance structures, inconsistent regulations, and lack of harmonized policies across African nations hinder the implementation of standardized soil pollution monitoring systems [191]. Fragmented and inconsistent environmental regulations across countries limit the enforcement of pollution controls and monitoring programs. There has been a clear recognition of gaps in the existing legal instruments and infrastructure related to the control, management, and remediation of soil pollution in several Sub-Saharan African (SSA) countries in that they are weak, shallow, and non-specific, making them ineffective in safeguarding environmental health. For example, articles 28 and 29 of the Environment and Natural Resources on Management of Chemicals protocol for the East African Community, the 1997 National Environmental Policy and Environmental Management Act of 2004 for Tanzania [52], the 1999 Constitution of the Federal Republic of Nigeria for Environmental policy aimed at protecting and safeguarding air, water, land wildlife, and forests are among some of the general and shallow legal frameworks in place. This is also typical for Zambia, where there is limited specificity regarding the management, control, and remediation of soil pollution [192].

5.2. Infrastructural, Financial, and Technological Limitations

Advanced technologies such as specialized equipment capable of detecting some of the soil pollutants that may have devastating effects within the environment and the ecosystem functions even at trace levels (e.g., most heavy metals and radioactive isotopes), programming within the GIS environment, remote sensing, and digital soil mapping are underutilized due to limited infrastructure, high costs, and limited expertise [191]. This implies that many countries rely on outdated methods that result in incomplete or inaccurate mapping products. Testing soil samples for pollutant quantification presents is costly, since most samples must be shipped to competent and well-equipped laboratories abroad due to limited local capacities. The region only has a limited number of laboratories that have the capacity to analyze samples for a wide range of contaminants. Most of the laboratories in the region are restricted to analyses for elements and trace elements. This leads to either inconclusive results or otherwise complicated and costly logistical arrangements to split samples for dispatch to multiple laboratories [52]. Many countries lack sufficient funding for large-scale soil pollution studies and mapping projects. Mapping soil pollution requires significant funding for fieldwork, laboratory analysis, technology acquisition, and capacity building. Unfortunately, many African nations prioritize other development sectors over environmental monitoring [193].

5.3. Data Availability and Accessibility

Many countries lack baseline data on soil properties, pollution levels, and sources. This makes it difficult to initiate detailed mapping. Africa lacks comprehensive, georeferenced soil data for many regions. This scarcity is due to limited soil sampling, lack of historical records, and inadequate investment in soil pollution monitoring [194]. In cases where the data exists, it is often inaccessible due to institutional barriers, lack of data-sharing agreements, and non-compliance with FAIR (findable, accessible, interoperable, and reusable) data principles [195]. This results in difficulties in collaborative research and regional initiatives, redundancy in data collection efforts, difficulty in establishing pollution baselines, limited ability to assess spatial and temporal trends, and challenges in prioritizing areas for remediation. Most National Soil Information Systems (SISs) for SSA do not have data and information on the fate of soil contaminants and their related toxicological effects integrated in their databases [196].

This state of pollution data inadequacies limits countries’ capacities to devise and implement effective plans and policies to curb soil pollution. Some studies, however, have highlighted the extent of pollution problems including the identification of specific sources contributing to soil contamination [52].

5.4. Climate Variability and Land Use Change

Africa’s varied climatic zones and the rapidly changing land use patterns significantly influence the deposition and spread of soil pollutants, making it difficult to adopt a uniform approach to soil pollution mapping across the continent [1]. Climate change contributes to soil contamination, primarily by influencing pollutant exposure and transport pathways due to changes in precipitation patterns, evaporation, runoff, and degradation rate. Climate change also brings about modifications in soil conditions, such as moisture and temperature regimes, soil reaction (pH), redox potential, and nutrient concentration, which significantly influence the dynamics and speciation of pollutants through various mechanisms, including adsorption–desorption and dissolution–precipitation [197]. Additionally, climate change intensifies extreme events, such as floods, which can accelerate erosion and surface runoff, causing significant changes in contaminant concentration, fate, and distribution [198]. According to some studies, heavy rainfall can significantly spread contaminants both horizontally and vertically into deeper soil horizons, making their detection and mapping difficult [199].

In general, rising temperatures, particularly due to climate change, can lead to the volatilization of major classes of globally significant toxic substances, including certain pesticides, such as organochlorines, as well as herbicides, fungicides, and other persistent organic pollutants (POPs) [200]. These processes add complexity to the accurate mapping of these pollutants, coupled with the fact that these pollutants vary and behave differently across different agro–ecological regions [197].

6. Recommendations for Advancing Soil Pollution Mapping and Management in Africa

Effective soil pollution mapping and management in Africa require improvements in data, capacity, infrastructure, and collaboration. Future initiatives should adopt robust sampling designs, include multiple soil depths and horizons, and follow standardized methods, such as FAO guidelines to ensure data quality and comparability [201,202,203]. Alongside data, building capacity is essential. Many African countries face shortages of trained professionals, equipment, and financial support. Investments in training, research, and awareness programs, supported by initiatives such as the Global Soil Partnership and Soil Doctors Program, can address these gaps and enhance national monitoring capacity [204].

Equally important is the development of national soil pollution databases and policies, which are largely absent across the continent. Establishing dedicated national databases would improve tracking of pollution trends, guide remediation, and inform policy. On a broader scale, systems such as GLOSIS demonstrate the potential of harmonized, transboundary soil information platforms but also highlight the urgent need to integrate soil pollution data [205,206,207].

Finally, public awareness and regional cooperation are key to long-term success. Educating communities and policymakers about soil pollution risks fosters sustainable practices and informed decision-making. Regional collaboration, through initiatives such as INSII or partnerships with international agencies, can facilitate knowledge sharing, standardization of monitoring protocols, and joint research [207]. Establishing regional monitoring centers would further enhance the capacity to track and mitigate pollution across borders. Together, these strategies can strengthen Africa’s ability to address soil pollution effectively.

7. Conclusions

Soil pollution is a critical environmental issue in Africa, posing significant threats to soil health, agricultural productivity, and ecosystem sustainability. Despite numerous studies on soil pollution, existing maps are often localized, inconsistent, and lack the necessary detail to inform effective decision-making. Heavy metals have been the primary focus of mapping efforts, while other pollutants, such as pesticides and persistent organic pollutants, remain underrepresented. The main challenges hindering comprehensive soil pollution mapping include fragmented, localized and inconsistent data, technological and institutional limitations, insufficient political commitment, and the impacts of climate variability. To enhance soil pollution mapping across Africa, it is essential to integrate the available data into a usable framework, strengthen data collection efforts, and develop national soil pollution databases to ensure reliable and spatially comprehensive information. Capacity building in soil pollution monitoring and mapping should be prioritized, alongside increased investment in research and technology. Additionally, fostering regional cooperation and raising public awareness can help drive coordinated efforts toward sustainable soil management. By addressing these gaps and challenges, Africa can improve its soil pollution mapping frameworks, ultimately supporting better land management and long-term environmental sustainability. With regard to the limitations of this study, we relied on literature that was published in English and French. It is therefore possible that we may have excluded relevant works published in other languages, such as Portuguese (Mozambique, Angola), Arabic (Morocco, Egypt, Algeria, Tunisia), or other local languages. Future works should therefore consider papers published in these languages. In addition, in selecting the different soil pollutants, this study did not consider emerging contaminants such as antibiotics and nanomaterials. Future studies should equally focus on these pollutants.