Respiration and carbon dynamics of free-living and particle-attached bacteria in coastal waters of NE Pacific

Cui Guo1, Ying Ke1 and Hongbin Liu2, (1)Hong Kong University of Science and Technology, Division of Life Science, (2)Hong Kong University of Science and Technology, Hong Kong, Hong Kong
Abstract:
Bacterial respiration (BR) rates are fundamental to understand the role of bacteria in carbon flow in aquatic ecosystem, and therefore it is critical to obtain reliable measurements. Prefiltration- (mostly 1-3μm) and dark-incubation- (mostly 24 h) based direct measurements of oxygen consumption have been the most commonly used method for BR. However, the prefiltration procedure and long incubation time may cause change of the bacterial abundance and structure, leading to inaccurate measurements. In this study, by measuring bacterial abundance, production and respiration of both particle-attached (PA) and free-living (FL) bacteria at two contrasting site in coastal NE Pacific from Nov 2014 to Mar 2015, we found that the 24 h growth rate of FL bacteria in the traditional BR incubations were significantly higher for ~30% and ~54% than those obtained for FL and total (FL+PA) bacteria in unfiltered incubations respectively, suggesting removal of protist grazers could cause a significant biomass accumulation during 24 h incubation than the in situ condition. This biomass overestimation resulted in ~40% (±12%) overestimation of measured FL BR rates compared with the corrected in situ FL BR. Nevertheless, for the corrected in situ total BR, the rates were overestimated by traditional method in three measurements over nine for 6-46%, and were underestimated in the rest six measurements for 7-67%. Interestingly, those underestimations were attributed to the ignorance of PA bacteria due to prefiltration, which had larger cell size than the FL bacteria, accounted for ~19% (±16%) in total bacterial abundance, and contributed to ~50% (±19%) of total bacterial production. The average bacterial growth efficiency calculated by comparable 24 h integrated bacterial production and respiration was 0.42 (±0.24). Our results confirmed two major flaws in the current BR methodology, i.e., 1) it only measures the respiration of FL bacteria, and 2) the removal of grazers causing dramatic increase of bacteria abundance during the 24 h incubation, and attempted to provide a solution for correction. Although the problems causes magnificent overestimate of the respiration of the FL bacteria, it could either overestimate or underestimate the total BR due to the proportion of PA bacteria that was not measured in the standard BR methodology.