Mechanisms driving Antarctic microbial community responses to ocean acidification: a network modelling approach

详细信息    查看全文

• 作者：Roshni C. Subramaniam ; Jessica Melbourne-Thomas ; Andrew T. Davidson…
• 关键词：Model ; CO2 ; Iron ; Antarctic microbial community
• 刊名：Polar Biology
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：40
• 期：3
• 页码：727-734
• 全文大小：
• 刊物类别：Biomedical and Life Sciences
• 刊物主题：Ecology; Oceanography; Microbiology; Plant Sciences; Zoology;
• 出版者：Springer Berlin Heidelberg
• ISSN：1432-2056
• 卷排序：40

文摘

Rising atmospheric CO₂ concentrations and the subsequent changes to ocean chemistry may have pronounced effects on marine microbial communities, particularly for the cold Southern Ocean. Changes to the microbial community in this region could affect the way nutrients are cycled, impact the efficiency of carbon drawdown, and cause shifts in food supply to higher trophic levels. Increased CO₂ could affect the bioavailability of iron to phytoplankton. Fertilisation experiments show that iron can influence phytoplankton community composition, favouring large phytoplankton species in iron-replete conditions. The potential interactive effects of CO₂ and iron bioavailability are currently poorly understood but are likely to be important in determining CO₂-induced changes to the microbial community. We employ a qualitative network modelling approach to evaluate alternative hypotheses regarding the effects of elevated CO₂ on Antarctic microbial communities in incubation experiments. We used a sequential approach to model development and testing, where we first formulated a base model for microbial community interactions, and then sequentially added direct and indirect effects of elevated CO₂ on particular groups. We found that model simulations were most consistent with observations from incubation experiments when we assumed an indirect effect of CO₂ on phytoplankton. In particular, when we assumed a negative effect of elevated CO₂ on the uptake of iron by large phytoplankton, as a result of a decrease in iron bioavailability. Our findings show that qualitative network models can be used to test hypotheses relating to results from experimental studies, and help identify key processes to target in future studies.