Sainz-López, N.; Boski, T., and Sampath, D.M.R., 2019. Fleur de sel composition and production: Analysis and numerical simulation in an artisanal saltern. Journal of Coastal Research, 35(6), 1200–1214. Coconut Creek (Florida), ISSN 0749-0208.
This study is a first approach to modelling of fleur de sel production, aimed at increasing its efficiency and contributing to the scarce literature on the topic. Quantitative forecasting of daily production of fleur de sel was applied to an artisanal solar pond unit in the environmentally protected area of Castro Marim, SE Portugal. The numerical model was based on simulations of the evaporation process, taking into account the effect of reduced vapour pressure of the brine solution. The controlling variables chosen as input parameters to the forecast model were brine temperature, brine concentration, harvesting efficiency, albedo, incoming solar radiation, precipitation, air relative humidity, air temperature, atmospheric pressure, day of the year, wind direction, and wind speed. Production predicted by the model was tested against actual production in two crystallisers in the years 2015 and 2017. The statistically evaluated match between the estimated and actual production was highly significant with a mean R² of 0.8 and overall error of estimation was 14.5%. The chemical composition of nine samples of fleur de sel was analysed, showing the temporal evolution of several components during the harvesting period. A decrease of NaCl content from 96% to 87% and an increase by one order of magnitude of Mg, S, K, Br, and As were observed. The range of Ca, Si, Al, and Sr contents was 0.12–0.65%, 0.1–0.85%, 0.08–0.3%, and 0.009–0.013%, respectively. Ba, Fe, Mn, Mo, Pb, and Sn contents had the following ranges: 0.051–0.145, 1.2–5.5, 0.44–1.66, 0.03–0.05, <0.05–0.55, and <0.05–0.1 mg/kg, respectively. The following elements were below the detection limit and below the limits of the Codex Alimentarius: Cd, Th, U, Cu, Cr, Co, Ni, V, Bi, Zn, Rb, and Hg.