Changes in phytoplankton assemblages may significantly alter elemental cycles. However, the differing contributions of phytoplankton functional types to the biological pumps have not been explored. This study aims to evaluate the sinking process of phytoplankton to the bathypelagic layer at the functional-type level. We collected sinking particles using sediment traps moored at 387 m and 890 m depths from June to August 2022 in the Sea of Japan and morphologically divided them into aggregates and fecal pellets (ellipsoidal, cylindrical, spherical, and tabular). The carbon flux of sinking particle types was measured, and the phytoplankton assemblages in every sinking particle type were investigated with 16S rRNA gene amplicon sequencing. The proportion of the phytoplankton-origin amplicon sequence variant (ASV) per total 16S gene sequence reads numbers in aggregates was 5.76 ± 0.496% (median ± interquartile range, n = 6) at 890 m, which was significantly lower than that in the ellipsoidal fecal pellets (8.96 ± 6.00%, n = 64) at 890 m depth. The number and sizes of the ellipsoidal pellets, considered appendicularian-origin, significantly increased with depth, and bigger ellipsoidal pellets were richer in phytoplankton ASVs. Diatom Chaetocerotales were the dominant phytoplankton group in each sample type and depth, except in the ellipsoidal and cylindrical fecal pellets at 890 m depth, where cyanobacteria Synechococcus was dominant. This suggests that phytoplankton, including Synechococcus, is effectively transported to the bathypelagic layer via the mesopelagic appendicularians repackaging process, while diatom Chaetocerotales effectively sink, regardless of the sinking processes.

Biological dinitrogen (N2) fixation is an important new nitrogen source in oligotrophic subtropical oceans. In numerical model studies, the east-west gradient of iron deposition as atmospheric Asian dust strongly affects the zonal distribution of N2 fixation activity in the North Pacific, but the in-situ relationship at a basin-scale is not well examined. We examined the trans-Pacific longitudinal variation in N2 fixation activity from 120°W to 137°E at 23°N in summer with environmental parameters that potentially influence diazotrophy. The dissolved inorganic iron concentration in surface water was consistently low (<0.4 nM) throughout the study area. The modelled deposition flux of iron as atmospheric dust (dust-Fe) largely increased westward, whereas labile phosphorus (phosphate and labile phosphoric monoesters) in the surface water decreased westward. N2 fixation varied between 34.6–298 µmol N m-2 day-1 and was high (>200 µmol m-2 day-1) in the central area (150–180°W), where both dust-Fe input and the phosphorus stock were in intermediate ranges. The rates of N2 fixation showed an increasing trend with dust-Fe input in the eastern and western parts of 180°, indicating that increasing dust input enhanced N2 fixation activity. However, compared with that of the eastern region, the effect of enhancement on activity was smaller in the western region, where phosphate concentration in the euphotic zone was low (<50 nM), presumably due to the higher iron requirement to utilize organic phosphorus. Our data show that phosphorus availability substantially controls the longitudinal distribution of N2 fixation through co-limitation with iron in the subtropical North Pacific.