Health risk assessment and the application of CF-PMF: a pollution assessment–based receptor model in an urban soil

Purpose This study was carried out to assess human health risk exposure, to apply a novel pollution assessment–based receptor model CF-PMF (contamination factor-positive matrix factorization), and to estimate the extent of contamination across seven cities in the Frydek-Mistek district. Nevertheless, the impact of agricultural production and industrial activities on urban soil and the livelihood of the indigenous peoples in the study area as well as the source contribution of the individual PTEs is unknown. Methods This study collected 49 soil samples across seven towns in the Frydek-Mistek district, which are primarily agricultural and industrially oriented urbanized communities. The samples were air-dried, and the potentially toxic elemental (PTEs) (i.e., Pb, As, Cr, Ni, Mn, Cu, and Zn) concentrations measured using portable x-ray fluorescence. Results Nemerow Pollution index and modified contamination degree indicated that the urban contamination levels were between low and moderate contamination level with a few cases of high contamination levels. The degree of contamination and the contamination factor showed varying levels of contamination for PTEs, with a high level of contamination and a low to high level of contamination, respectively. PTEs displayed a low to high pattern of spatial distribution in urban soil around Trinec, Bystice, Hrcava, and Harirov. The source of the PTEs was detected using principal component analysis, and the source apportionment of the PTEs was further assessed using CF-PMF (contamination factor-positive matrix factorization). Comparison of the CF-PMF receptor model and the EPA-PMF receptor model revealed that the novel receptor model performed better. The root mean square error (RMSE) and the mean absolute error (MAE) of the new receptor model marginal errors reduced significantly. RMSE and MAE for the CF-PMF receptor model for all the PTEs for instance As, Cr, Cu, Mn, Ni, Pb, and Zn are 11.56, 97.85, 17.30, 527.26, 37.16, 32.12, and 68.02 (RMSE) and 11.58,95.00, 17.26, 520.85, 37.04, 32.13, and 68.03 (MAE) were lesser than the EPA.PMF receptor model respectively. Health risk computed indicated that there was no potential carcinogenic and non-carcinogenic risk being exposed to the people living within the study area. Conclusion We propose using the novel receptor model CF-PMF because its output has shown to be optimal with minimal error and improved efficiency when compared to the parent model EPA-PMF. In general, continuous introduction of agro-related inputs and other anthropogenic activities surges PTEs levels in urban soils. Thus, constructive yet efficient steps, appropriate control, and mitigation measures are required to abate pollution sources that may be sowed to the soil.