Community structure ecosystem function relationships in the Congo Basin methane cycle depend on the physiological scale of function
Belowground ecosystem processes can be highly variable and difficult to predict using microbial community data. Here we argue that this stems from at least three issues: 1) complex covariance structure of samples (with environmental conditions or spatial proximity) can make distinguishing biotic drivers a challenge, 2) communities can control ecosystem processes through multiple mechanisms, making the identification of these controls a challenge and 3) ecosystem function assessments can be
... ssments can be broad in physiological scale, encapsulating multiple processes with unique microbially mediated controls. We suggest that these problems can be overcome by addressing environmental variability and sample covariance structure, separating microbial community measurements into distinct attributes that represent putative controls, and resolving functions to finer physiological scales. We test these assertions using methane (CH4)-cycling processes in soil samples collected along a wetland-to-upland habitat gradient in the Congo Basin. We perform our measurements of function under controlled laboratory conditions and include environmental covariates in statistical analyses to aid in identifying biotic drivers. We divide measurements of microbial communities into four attributes (abundance, activity, composition, and diversity) that represent different forms of community control. Lastly, our process measurements differ in physiological scale, including broader scales (gross methanogenesis and methanotrophy) that involve more mediating groups, to finer scales (hydrogenotrophic methanogenesis and high-affinity CH4 oxidation) with fewer mediating groups. We observed that finer scale processes can be more readily predicted from microbial community structure than broader scale processes. In addition, the nature of those relationships differed, with broad processes limited by abundance while fine-scale processes were associated with diversity and composition. These findings demonstrate the importance of carefully defining the physiological scale of ecosystem function and performing community measurements that represent the range of possible controls on ecosystem processes.