Technical note: Measurements of fluorescent dissolved organic matter (FDOM) in seawater (filter blanks, pore sizes, and storage)

Abstract. Fluorescent dissolved organic matter (FDOM) provides crucial information regarding the sources and characteristics of dissolved organic matter (DOM) in oceans. However, results from FDOM measurement can depend on filter blanks, pore sizes, and sample storage. To develop more reliable methods for FDOM measurements, we examined uncertainties associated with different preparation methods for seawater samples. Three primary components were identified from these samples using parallel factor analysis: terrestrial humic-like peak (C peak), marine humic-like peak (M peak), and protein-like peak (T peak). Relatively high blank values were observed when samples were filtered through a pre-combusted glass fiber filter (Whatman, borosilicate, 0.7 µm, 47 mm) and a membrane filter (Whatman, mixed cellulose ester, 0.2 µm, 47 mm) without pre-cleaning. These blank values were negligible when both filters were washed with 5 mL of 0.1 M HCl (∼0.29 mL cm−2) or 20 mL of distilled water (∼1.16 mL cm−2). The effects of different filter pore sizes were not observed for the C and M peaks, but lower T-peak values were observed for filtered samples relative to unfiltered samples. During storage, C and M peaks showed consistent results for 21 d (8 % ± 3 %) when kept in pre-combusted amber glass vials in a refrigerator or a freezer. In contrast, clear changes were observed in samples stored at room temperature after 5 d. Thus, reliable C and M peaks can be obtained from unfiltered or filtered samples stored in a refrigerator or freezer for up to 3 weeks. However, T-peak intensity decreased rapidly in both filtered (15 %–50 %) and unfiltered samples (10 %–40 %) within 5 d, indicating the influence of significant biological and abiotic processes. Therefore, our results suggest that careful sample filtration, storage, and blank controls are necessary for T-peak measurements.