Open data for honey adulteration and authentication: dataset catalogue

A catalogue of datasets for honey adulteration and authentication across five analytical modalities: Fourier-transform infrared (FTIR) spectroscopy, hyperspectral imaging (HSI), near-infrared, visible/near-infrared and UV-Vis spectroscopy (NIR/VIS-NIR), physicochemical profiling, and multimodal fusion of these. It records not how well each method performs, but whether the data behind each study can be found, obtained and reused. The catalogue holds 83 dataset entries from 82 unique studies published between 1996 and 2026. Seventy-three entries concern honey; ten concern other foods (fruit purée, coffee, olive oil, meat, wine, pistachio, apple, wheat, grape and prepared dishes) and are included as deposition templates, because they release data in forms that honey datasets lack, such as raw hypercubes and openly licensed spectra. For each entry the catalogue records the food matrix, task, samples, spectral range, origin, year, repository, data link, associated article, access status with the evidence it was based on, data licence, what is released, file format, raw hypercube status (HSI), and five indicative FAIR proxies (Findable, data identifier, Accessible, Interoperable, Reusable). Access is coded in four categories: Public (21 entries), On request (23), Restricted (2) and No route stated (37). Twelve public datasets hold honey measurements. Records came from a census of data repositories with one hop of backward citation chaining, searches of seven data repositories and OpenAlex, a Scopus search (813 records, each with its screening decision), and targeted web and reference-list searching. Of 157 records assessed in full, 83 were included and 74 were excluded with a stated reason. Access statuses were checked between 11 and 31 October 2026; repository records change, so a status may have changed since. The catalogue is a structured, non-exhaustive survey, not a systematic review, and shares computed from it describe the catalogued entries rather than the whole literature. The FAIR proxies are indicative and are not a FAIR compliance assessment. Files:- Honey_Food_Authentication_Dataset_Catalogue_v4.xlsx: the catalogue workbook (Read Me, All Datasets, Summary by Modality with live formulas, Excluded, Search Log, Research Gaps, Raw HSI Formats Guide)- csv/: plain UTF-8 copies of the main sheets- Scopus_screening.csv: the screening decision for each of the 813 Scopus records- figures.py and figures/: a script that regenerates the four figures of the associated review from the workbook and the Scopus file, and the figures as PDF- README.md: definitions of every access category and FAIR proxy, and the changes from v3 Version v4 supersedes v3 (10.5281/zenodo.21818940). Entries grew from 38 to 83, access is coded in four categories instead of three, all records that were pending in v3 are resolved, and the research gaps were revised to seven, each paired with the minimum deposition that would close it.