Deep sea circulation of particulate organic carbon in the Japan Sea

Seasonal and spatial variations of particulate organic carbon (POC) flux were observed with sediment traps at three sites in the Japan Sea (western and eastern Japan Basin and Yamato Basin). In order to investigate the transport processes of POC, radiocarbon (¹⁴C) measurements were also carried out. Annual mean POC flux at 1 km depth was 30.7 mg m⁻²day⁻¹ in the western Japan Basin, 12.0 mg m⁻²day⁻¹ in the eastern Japan Basin and 23.8 mg m⁻²day⁻¹ in the Yamato Basin. At all stations, notably higher POC flux was observed in spring (March–May), indicating biological production and rapid sinking of POC in this season. Sinking POC in the high flux season showed modern Δ¹⁴C values (>0‰) and aged POC (Δ¹⁴C < −40‰) was observed in winter (December–January). The Δ¹⁴C values in sinking POC were negatively correlated with aluminum concentration, indicating that Δ¹⁴C is strongly related to the lateral supply of lithogenic materials. The Δ¹⁴C values also showed correlations with excess manganese (Mn_xs) concentrations in sinking particles. The Δ¹⁴C-Mn_xs relationship suggested that (1) the majority of the aged POC was advected by bottom currents and incorporated into sinking particles, and (2) some of the aged POC might be supplied from the sea surface at the trap site as part of terrestrial POC. From the difference in the Δ¹⁴C-Mn_xs relationships between the Japan Basin and the Yamato Basin, we consider that basin-scale transport processes of POC occur in the Japan Sea.     

Throughfall partitioning by trees

Although we know that rainfall interception (the rain caught, stored, and evaporated from aboveground vegetative surfaces and ground litter) is affected by rain and throughfall drop size, what was unknown until now is the relative proportion of each throughfall type (free throughfall, splash throughfall, canopy drip) beneath coniferous and broadleaved trees. Based on a multinational data set of >120 million throughfall drops, we found that the type, number, and volume of throughfall drops are different between coniferous and broadleaved tree species, leaf states, and timing within rain events. Compared with leafed broadleaved trees, conifers had a lower percentage of canopy drip (51% vs. 69% with respect to total throughfall volume) and slightly smaller diameter splash throughfall and canopy drip. Canopy drip from leafless broadleaved trees consisted of fewer and smaller diameter drops (D_{50_DR}, 50th cumulative drop volume percentile for canopy drip, of 2.24 mm) than leafed broadleaved trees (D_{50_DR} of 4.32 mm). Canopy drip was much larger in diameter under woody drip points (D_{50_DR} of 5.92 mm) than leafed broadleaved trees. Based on throughfall volume, the percentage of canopy drip was significantly different between conifers, leafed broadleaved trees, leafless broadleaved trees, and woody surface drip points (p ranged from <0.001 to 0.005). These findings are partly attributable to differences in canopy structure and plant surface characteristics between plant functional types and canopy state (leaf, leafless), among other factors. Hence, our results demonstrating the importance of drop‐size‐dependent partitioning between coniferous and broadleaved tree species could be useful to those requiring more detailed information on throughfall fluxes to the forest floor.