A dataset of PFASs concentrations, bioaccumulation and translocation, and dietary risk assessment in farmland–rice systems around a typical fluorochemical industrial park

Weak‑anion‑exchange solid‑phase extraction (WAX‑SPE) was used for enrichment and purification, and 37 PFAS monomers were quantitatively determined by ultra‑performance liquid chromatography‑tandem mass spectrometry (UPLC‑MS/MS). Quality control was performed in accordance with US EPA Method 1633. Isotope‑dilution quantification was adopted.

Fluoropolymer‑containing materials were avoided throughout the experiment, and field blanks, transport blanks and matrix spike samples were prepared. Based on the measured PFAS concentrations, bioaccumulation factors (BAF) and translocation factors (TF) were calculated. Dietary exposure risk assessment for different populations was further implemented to obtain the estimated daily intake (EDI), hazard quotient (HQ) and hazard index (HI).This dataset consists of discrete field observation data.

Sampling was conducted once in November 2023 at the rice mature‑harvest stage, covering five spatial sampling sites. It is not a remote‑sensing gridded dataset and has no temporal or spatial resolution. All data are stored in a single Excel workbook (.xlsx) containing four worksheets.

Sheet1 “Concentrations of PFASs in different rice tissues”: five sampling sites (ST1‑ST5) × five tissue types (brown rice, white rice, rice root, rice stem, rice leaf) are used as column identifiers, and PFAS monomers serve as row identifiers. The former block (conc) contains measured concentrations, and the latter block (percent) represents compositional proportions (%). Concentrations in rice tissues are expressed in ng·g⁻¹ (dry weight).

Sheet2 “Concentrations of PFASs in soil and irrigation water”: it comprises four sub‑blocks: soil conc, soil percent, water conc and water percent. Five sampling sites (ST1‑ST5) are set as columns, and PFAS monomers serve as row identifiers. Soil concentrations are in ng·g⁻¹ (dry weight), and irrigation‑water concentrations are in µg·L⁻¹.

Sheet3 “Bioaccumulation factors (BAF) and translocation factors (TF)”: divided into BAF and TF sub‑blocks. Five sampling sites (ST1‑ST5) are columns, and rows are organized by “compound × rice tissue” (BR for brown rice, WR for white rice, Root for root, Shoot for stem, Leaf for leaf). It covers PFASs commonly detected in both soil and rice.

BAF and TF are dimensionless parameters and are presented as log₁₀‑transformed values in the table. Sheet4 “Estimated daily intake (EDI), hazard quotient (HQ) and hazard index (HI)”: organized into EDI, HQ and HI sub‑blocks. EDI and HQ datasets cover five sampling sites, urban/rural populations, and four age groups: 2‑6, 7‑18, 19‑45, >45 years old.

Each age group is further divided into brown rice (BR) and white rice (WR). Rows represent 33 PFAS monomers, and columns are grouped by “residential area × age × rice tissue” (BR for brown rice, WR for white rice). EDI is in ng·kg⁻¹·d⁻¹; HQ and HI are dimensionless.

Population exposure parameters (daily rice intake, body weight) for risk assessment were derived from the China Health and Nutrition Survey (CHNS), Chinese Food Composition Tables, and the follow‑up study on nutrition and health status of residents in Jiangsu Province. The reference dose (RfD) values of PFASs were obtained from review documents published by the Hawaii Department of Health of the United States. For substances without available RfD values, the structural‑analog substitution method was adopted for assignment.Regarding data missing and uncertainty: values below the limit of quantification (LOQ) were assigned as LOQ/2, and the marker <LOQ> indicates analytes below the quantification limit.

Non‑detected compounds in Sheet1 are denoted as 0. Uncertainty sources include spatial heterogeneity during field sampling, matrix effects in sample pretreatment, and instrumental fluctuations of UPLC‑MS/MS. The matrix spike‑recovery ranges for PFASs are 61%‑167% for water matrix, 37%‑168% for soil matrix, and 56%‑233% for plant matrix.

Detailed LOQ and recovery data are listed in Table 1 of the corresponding published paper. No complete sample loss occurred in this dataset; only partial compounds had concentrations below LOQ. The files are in universal .xlsx format and can be directly opened and read by Microsoft Excel, WPS Office and Origin.