The team’s target list reads like a periodic table of environmental concern: arsenic, beryllium, cadmium, cobalt, chromium, copper, iron, manganese, nickel, lead, vanadium and zinc. Each of these elements can arrive in soil by more than one route. Some, such as vanadium and much of the iron, typically reflect the mineral composition of the parent geological material from which the soil formed. Others, notably copper, lead, zinc and chromium, are classic tracers of human activity — the residue of historical metalworking and manufacturing, of leaded petrol and legacy paints, of brake pads and tyre wear, of coal ash and the general churning of made-ground that defines urban land. Arsenic sits awkwardly in between: it can be geogenic, tied to clays and iron oxides, yet it is also a well-known industrial contaminant. Disentangling these origins is the central analytical challenge of any urban soil survey, and it is precisely the challenge the Leicester study set out to address with a deliberately structured sampling design.

Rather than chasing individual hotspots, the researchers adopted what they describe as an area-level geochemical screening approach. Eighteen urban parks and open spaces within Leicester and eight rural areas across Leicestershire were surveyed. Within each area, the team collected 850 individual field increments — small, spatially distributed samples of topsoil — and then pooled and homogenised them into a single area-level composite. From each composite, two separately identified laboratory samples were prepared, yielding 52 samples in total. These were then subsampled again for independent analytical processing, producing 102 valid analytical subsamples. The logic of this layered design is statistical as much as chemical: by averaging many increments, the composite captures the diffuse, average geochemical condition of an area rather than the noise of any single hotspot, and by duplicating laboratory samples and analytical subsamples the team could quantify variability introduced after the field stage. It is a design philosophy that trades fine-grained spatial resolution for a robust, defensible picture of area-level conditions — an appropriate trade-off for screening purposes.

Chemically, the workflow followed established protocols for pseudo-total digestion. Soil samples were attacked with a mixture of nitric and hydrochloric acids — the HNO₃/HCl combination known in the trade as aqua regia-style digestion — which dissolves the environmentally reactive fraction of the metals bound to organic matter, oxides and mineral surfaces, though not the silicates locked deep in unweathered grains. The digests were then analysed by inductively coupled plasma mass spectrometry, or ICP-MS, a technique capable of detecting elements at part-per-billion levels by ionising the sample in an argon plasma and sorting the resulting ions by their mass-to-charge ratio. Iron, present at much higher concentrations, was measured instead by atomic absorption spectrometry, which is well suited to major elements. Notably, the team treated cadmium and chromium as left-censored variables — measurements falling below the limit of detection were handled with formal censored-data methods rather than the common but statistically dubious practice of substituting arbitrary values such as zero or half the detection limit. This matters: with trace contaminants, how nondetects are handled can materially shift calculated means, correlations and risk estimates.

The headline geochemical finding is a clean urban–rural split. Urban Leicester topsoils were enriched in chromium, copper and zinc relative to their rural counterparts — a pattern consistent with diffuse urban inputs, traffic-related emissions and the accumulated legacy of historical land use. Copper and zinc, in particular, are signature elements of the modern road environment: brake pad wear releases copper, while zinc comes from tyre wear particles, galvanised surfaces and assorted urban debris. Chromium’s urban excess points toward industrial sources, including historical metal finishing and coal combustion. Rural topsoils, by contrast, showed higher arsenic and vanadium, supporting the interpretation that in the countryside the soil’s mineral matrix — its parent material, clay content and iron and manganese oxides — exerts the dominant control on metal(loid) concentrations. In other words, the city’s soils remember its factories and its traffic, while the country’s soils remember its geology.

To quantify enrichment, the researchers calculated several standard geochemical indices, normalised against the regional background established by the British Geological Survey’s soil atlases. The pollution index, enrichment factor and pollution load index each compare measured concentrations against reference baselines, and all three pointed toward mild to moderate enrichment across the sampled areas. The clearest cumulative signal emerged in the south-west quadrant of the study area, and when concentrations were normalised to iron — a technique that attempts to cancel out natural mineralogical variation by expressing contaminants relative to a geogenic anchor element — copper and lead showed the strongest enrichment signatures. Iron normalisation is a well-established device in urban geochemistry because iron is abundant, largely immobile and overwhelmingly natural; a contaminant-to-iron ratio that rises above the regional background ratio is strong evidence of anthropogenic addition.

The multivariate statistics reinforced the same two-source story. Principal component analysis, a technique that condenses many correlated variables into a few independent axes of variation, together with Spearman rank correlation, separated the twelve elements into two coherent groups. One component bound iron, arsenic and vanadium — the matrix-related group, tracking the soil’s natural mineralogy. The other grouped chromium with copper, lead, nickel, cadmium and zinc — the enrichment-sensitive group, tracking human inputs. That the statistical structure of the dataset independently reproduces the urban–rural comparison gives the geochemical interpretation considerable weight: two different analytical lenses produce the same picture of natural versus anthropogenic control.

The most consequential part of the study is its screening-level health assessment, built on the framework of Category 4 Screening Levels developed for UK contaminated-land practice, combined with the US Environmental Protection Agency’s risk assessment methodology. The team computed hazard indices — the sum of hazard quotients across elements and exposure pathways — for hypothetical residents, workers and users of public-access land, using precautionary assumptions about soil ingestion rates and exposure frequencies. Under those deliberately conservative assumptions, hazard indices for child residents and child users of public-access spaces exceeded unity, meaning the estimated exposure surpassed the screening threshold, and arsenic was the main driver. Adult resident and worker hazard indices remained below unity. Lead, assessed with dedicated screening ratios appropriate to its particular toxicology and absence of a safe exposure threshold, remained below unity. Estimated lifetime cancer risks from the carcinogenic elements fell within the 10⁻⁶ to 10⁻⁴ range conventionally regarded as acceptable-to-tolerable in regulatory practice.

The authors are careful — appropriately so — about what these numbers mean and do not mean. A hazard index above unity under precautionary screening assumptions is a flag, not a diagnosis. It identifies areas and elements where a more refined, site-specific assessment would be justified, incorporating actual bioaccessibility of the metals in the soil matrix, real exposure scenarios and soil-specific parameters. The study is explicitly described as composite-derived geochemical screening evidence rather than a statutory contaminated-land determination under Part 2A of the UK environmental protection regime. That distinction matters for public communication: the results should inform prioritisation and further investigation, not alarm, and the fact that the exceedances arise mainly from arsenic — an element often elevated in English soils for geological reasons — underlines why regional normal background concentrations are an essential benchmark in these assessments.

What elevates the work beyond a local survey is its methodological transparency and openness. The full dataset, including raw and calculated concentrations, limits of detection and below-detection flags, has been released on the Open Science Framework under a Creative Commons Attribution licence, allowing other researchers to reanalyse, compare and build on the results. Leicester itself is a historically rich test case: a city shaped by boot and shoe manufacture, hosiery and engineering, with a post-industrial transformation from the 1970s onward and a geology of Mercia Mudstone and glacial deposits that varies across the district. As UK cities confront the twin pressures of densification and the conversion of post-industrial land to green infrastructure, cheap, rigorous, composite-based screening surveys of the kind demonstrated here could become a standard first line of defence — a way of knowing, in advance and at modest cost, what the ground beneath the playground is quietly keeping to itself.