Contrasting carbon cycling in the benthic food webs between riverfed, high-energy canyon and upper continental slope.

The Gaoping Submarine Canyon (GPSC) off Southwest Taiwan has been extensively studied due to its unique geology, the role of transferring terrestrial material to the deep sea, and diverse biological communities. However, there is a lack of understanding of carbon cycling across the sediment--water interface in the canyon. This study aims to fill the gap by utilizing the field data collected between 2014 and 2020 and the Linear Inverse Model (LIM) to reconstruct the benthic food web (i.e., carbon flows through different stocks) in the head of GPSC and the upper Gaoping slope (GS). The biotic and abiotic organic carbon (OC) stocks were significantly higher on the slope than in the canyon, except for the bacteria stock. The sediment oxygen utilization was similar between the two habitats, but the magnitude and distribution of the OC flow in the food web were distinctively different. Despite a significant input flux of ~1400 mg C m^-2 d^-1 in the canyon, 85% of the carbon flux exited the system, while 14% was buried. On the slope, 75% of the OC input (~105 mg C m^-2 d^-1) was buried, and only 11% exited the system. Bacteria processes play a major role in the carbon fluxes within the canyon. In contrast, the food web in the upper slope exhibited stronger interactions among metazoans, indicated by higher fluxes between meiofauna and macrofauna compartments. Network indices based on the LIM outputs showed that the canyon head had higher total system throughput (T..) and total system through flow (TST), indicating greater energy flowing through the system. In contrast, the slope had a significantly higher Finn cycling Index (FCI), average mutual information (AMI), and longer OC turnover time, suggesting a relatively more stable ecosystem with higher energy recycling. Due to sampling limitations, the present study only represents the benthic food web during the "dry" season. By integrating the field data into a food web model, this study provides valuable insight into the fates of OC cycling in an active submarine canyon, focusing on the often overlooked benthic communities. Future studies should include "wet" season sampling to reveal the effects of typhoons and monsoon rainfalls on OC cycling. [ABSTRACT FROM AUTHOR]