Growing concern has been expressed about uranium (U) accumulation in agricultural soils caused by the long-term application of mineral fertilizers. More than 80% of naturally occurring U transfers from phosphate rock (PR), the raw material used in mineral fertilizer production, to phosphorus (P) fertilizers. These fertilizers are then distributed on agricultural soils, where the U could accumulate over time and become a risk to the environment. The objective of this work was to review the reported content of U in P fertilizers, its potential dispersion in soils, and its uptake by plants in different countries in the world as reported in the literature. The articles for this systematic review were selected from the Scopus database published between 2003 and 2022. The preferred reporting items for systematic reviews and meta-analyses (PRISMA) protocol were used. A total of 54 articles were assessed based on the standard inclusion and exclusion criteria. U concentrations in P fertilizers, agricultural soil dissemination, and plant uptake for available data were obtained and assessed. In order to compare a set of related data from the collected articles, box and whisker plots showing the distribution of U in P fertilizers are presented by region. The results from the reviewed articles show that the U concentrations in P fertilizer were in the range of 0.1–653 mg kg⁻¹. Interestingly, Minjingu P fertilizers from Tanzania, which are used in six East African countries, showed the highest U concentrations (159 to 653 mg kg⁻¹, average 390 mg kg⁻¹). The reported U concentrations for these fertilizers are, in fact, comparable to those of conventional low-grade uranium deposits mined in Namibia and elsewhere. Additionally, approximately 96% of the reviewed articles indicate that fertilized soil has higher U concentrations than non-fertilized soils, hinting at a measurable effect of mineral fertilizer use. The review recommends U extraction during mineral fertilizer production so that potential environmental risks can be reduced and U resources that would otherwise be lost can be recovered and used to substitute conventional U mining elsewhere.

Long-term ingestion of water contaminated with naturally occurring radioactive material (NORM) may pose health risks. Water around the Geita Gold Mine in Tanzania was assessed by high-purity germanium gamma spectrometry to quantify the activity concentrations of ²²⁶Ra, ²³²Th, and ⁴⁰K, and computed age-stratified ingestion doses and risk indices were determined. The average activity concentrations were 57 mBq L⁻¹ for ²²⁶Ra and 5026 mBq L⁻¹ for ⁴⁰K, while the activity concentrations of ²³²Th were below the detection limit in all samples. The estimated adult fatal cancer risk ranged from 0.9 × 10⁻⁶ to 3.1 × 10⁻⁶ (mean 2.0 × 10⁻⁶). The excess lifetime hereditary effect ranged from 2.0 × 10⁻⁶ to 7.3 × 10⁻⁶ for males (average 4.5 × 10⁻⁶ ± 1.5 × 10⁻⁶) and 2.1 × 10⁻⁶ to 7.7 × 10⁻⁶ for females (average 4.8 × 10⁻⁶ ± 1.6 × 10⁻⁶). One-way ANOVA and Pearson correlations indicated significant spatial variation in activities and indices across sites and age groups. Under current conditions, waters appear to be radiologically safe. However, mine-adjacent hotspots warrant targeted surveillance. The obtained results provide a baseline for sound monitoring approaches at the Geita Gold Mine and other mines showing similar activity profiles.

The presence of radiotoxic uranium (U) in mineral fertilizers is of global concern. A pilot study was conducted in Tabora (Tanzania) to determine the release of U from three brands of phosphate fertilizers and its impact on soil bacteria. The experiment used three types of fertilizer: Minjingu Powder (MP), Nafaka Plus (NP), a mixed and granulated fertilizer made from Minjingu Phosphate Rock (MPR), and YaraMila Cereal (YC) fertilizer. There was also a control treatment that was not fertilized (NF). Alpha diversity and the R tool were used to analyze bacterial diversity in four samples within an average sequencing depth of 74,466 reads, using metrics like ASVs, Shannon index, and Chao1. The results showed that the number of amplicon sequence variants (ASVs) in the DNA from soil bacteria decreased, specifically to 400 ASVs, in the NP treatment, which was in line with the higher U concentration (3.93 mg kg⁻¹) in the soils. In contrast, the MP fertilizer treatment, associated with a lower U concentration (3.06 mg kg⁻¹) in soils, exhibited an increase in ASVs within the DNA of soil bacteria, reaching 795; the highest ASV value (822) was observed in the NF treatment. Higher amounts of U in the soil plots seemed to have resulted in more types of bacteria, with the Actinobacteriota phylum being the most common in all of the treatments. The NP (3.93 mg kg⁻³ U concentration) and MP (3.06 mg kg⁻³ U concentration) treatments were the only ones that showed Halobacteriota and Crenarchaeota phyla. Nonetheless, bacterial diversity may also account for the alterations in soil phosphorus and nitrogen following fertilizer application. The YaraMila Cereal treatment did not seem to be linked to any particular bacterial phylum. This means that in this study it did not have any measurable effect on the soil bacteria species compared to the MP and NP treatments.