Arsenic and cadmium raise concern in Bangladesh farm soils and crops

A new study of agricultural fields in Nagarkanda Upazila, Faridpur, Bangladesh, has found that arsenic concentrations in every one of fifteen sampled agricultural soils exceeded a cited permissible limit of 20 milligrams per kilogram, while cadmium emerged as the dominant driver of ecological risk in the same fields. The research, published in Discover Soil, paired soil samples with edible crops including rice, brinjal, spinach, cabbage, lentil, onion, radish, coriander leaf, onion grass and elephant ear, collected between October and December 2024 from nine villages across four unions of the upazila.

The researchers measured arsenic, lead, chromium and cadmium using flame atomic absorption spectrometry after acid digestion of oven-dried samples. Mean soil concentrations were 36.756 milligrams per kilogram for arsenic, 11.850 for lead, 64.933 for chromium and 0.373 for cadmium, with arsenic ranging from 21.730 to 46.520. In the edible crops, mean concentrations were far lower: 0.140 milligrams per kilogram for arsenic, 0.524 for lead, 3.675 for chromium and 0.0646 for cadmium. The highest soil arsenic, 46.520 milligrams per kilogram, was found in a rice field at Khuduzia, while the highest soil chromium and cadmium, 94.641 and 0.502 milligrams per kilogram, occurred in a rice field at Chagaldi.

Against the guideline values adopted by the study, all fifteen soil samples exceeded the arsenic limit of 20 milligrams per kilogram, whereas lead, chromium and cadmium remained below their respective soil limits of 100, 100 and 3 milligrams per kilogram. The gap between soil and crop concentrations reflects differences in metal retention, bioavailability and plant uptake. Plants can restrict excessive metal accumulation through immobilization, sequestration and antioxidant defense responses, and metal mobility is shaped by soil pH, organic matter, texture and cation-exchange processes, none of which were directly measured in this study.

Soil-to-plant transfer factors, calculated as the ratio of plant to soil concentration for each paired sample, were all below one. Cadmium showed the highest mean transfer factor at 0.179, followed by lead at 0.061, chromium at 0.060 and arsenic at just 0.004. This ranking is consistent with the generally higher mobility of cadmium compared with strongly sorbing metals such as lead and chromium, although the authors caution that transfer varies with crop species and soil properties and should not be treated as a fixed uptake coefficient.

The contamination indices told a consistent story. Using regional background values of 12.70 milligrams per kilogram for arsenic, 0.15 for cadmium, 67.30 for chromium and 21 for lead, the contamination factor was highest for arsenic, with a mean of 2.894, placing nine samples in the moderately contaminated class and six in the considerably contaminated class. Cadmium followed with a mean contamination factor of 2.484. Lead and chromium showed much lower mean values of 0.564 and 0.965. The geo-accumulation index similarly identified arsenic and cadmium as the primary enrichment concerns, with both metals falling in the unpolluted to moderately polluted range, while lead and chromium showed negative values across all samples, indicating concentrations within natural background variability.

The Pollution Load Index, a combined measure across the four metals, ranged from 0.857 to 2.020 with a mean of 1.353. Thirteen sites were classified as moderately polluted, one site, S-2 at Baghat, was heavily polluted with a value of 2.020, and one site, S-7 at Brammondanga, showed no pollution. The Potential Ecological Risk Index ranged from 101.60 to 168.77 with a mean of 133.05. Twelve sites fell in the low ecological risk category and three showed moderate risk, with none exceeding the high-risk threshold. Cadmium dominated the ecological risk calculation, contributing an average of about 99.35 of the total index, compared with 28.94 for arsenic, 2.82 for lead and 1.93 for chromium, a reflection of cadmium’s large toxic-response coefficient of 40 in the Hakanson model.

Dietary risk screening raised separate concerns. Using standard exposure equations with assumed ingestion rates of 120 grams per day for children and 250 grams per day for adults, the study estimated average daily doses and hazard quotients for vegetable and rice consumption. Arsenic hazard quotients exceeded the safety threshold of one in all scenarios, reaching 1.64 for children and 1.99 for adults from vegetables and 1.53 and 1.86 from rice. Chromium produced far higher values, up to 18.54 for adults consuming rice, but only because the analysis applied chromium(VI)-specific toxicity benchmarks to total chromium measured by atomic absorption, a deliberately conservative assumption that treats all measured chromium as the more toxic hexavalent form. The combined hazard index, which excluded lead because no oral reference dose was available, reached 20.71 for adult rice consumption, driven mainly by the chromium assumption.

The authors stress that these dietary estimates are screening values, not predictions of harm. Plant concentrations were measured on oven-dried material, while food standards and dietary intake are generally expressed on a fresh or consumed basis, and no moisture conversion, cooking-loss correction or local dietary survey data were available. The ingestion assumptions were retained from earlier studies for comparability rather than derived from Nagarkanda eating patterns. Carcinogenic risk estimates, which ranged up to 4.34 × 10⁻³ for adult rice consumption, again dominated by chromium, likewise depend on the conservative chromium(VI)-equivalent scenario and fall outside straightforward interpretation without speciation data.

Multivariate analysis added context to the spatial and chemical patterns. Principal component analysis showed that the first two components explained 96.8 percent of the total variance in the standardized dataset, with plant samples clustering separately from the more dispersed soil samples. Hierarchical clustering split the thirty observations into five groups, with all fifteen plant samples forming one tight cluster and the soils dividing into four groups of varying internal similarity, including a distinct pair formed by soil samples S-13 and S-14. Pearson correlations suggested moderate associations between plant arsenic and cadmium, plant lead and cadmium, and soil chromium and cadmium, but with only fifteen pairs and two variables departing from normality, these were treated as exploratory co-variation rather than evidence of shared sources or antagonistic uptake.

The study’s limitations are substantial and acknowledged in detail. Fifteen purposively selected sites across an area of roughly 192 square kilometers represent a single-season snapshot with no seasonal replication and no formal sample-size calculation. Archived quality-control records lacked calibration coefficients, blank results, certified reference material recoveries and detection limits, so compliance with the laboratory’s own acceptance criteria cannot be independently verified. Site-specific background soils were unavailable, forcing reliance on regional literature values, and the inverse distance weighting maps generated in ArcGIS could not be cross-validated. The authors position the work as a preliminary local screening baseline intended to guide denser, seasonally replicated monitoring, including chromium speciation, irrigation-water analysis and soil physicochemical characterization, rather than as standalone evidence for regulatory decisions.

Subject of Research: Heavy metal contamination and ecological and dietary health risks in agricultural soils and edible crops of Nagarkanda, Faridpur, Bangladesh

Article Title: Potential ecological and human health risks associated with heavy metals in agricultural soils and edible crops of Nagarkanda Bangladesh

Article References: Hossan, M. R., Akter, S., Brahma, P. P., & Islam, R. (2026). Potential ecological and human health risks associated with heavy metals in agricultural soils and edible crops of Nagarkanda Bangladesh. Discover Soil, 3(1), Article 201. https://doi.org/10.1007/s44378-026-00349-z

Image Credits: AI Generated

DOI: 10.1007/s44378-026-00349-z

Keywords: heavy metals, arsenic, cadmium, agricultural soil, food safety, Bangladesh, soil contamination, ecological risk, transfer factor, health risk assessment, rice, vegetables