Symbiotic relationships between N 2 -fixing prokaryotes and their autotrophic hosts are essential in nitrogen (N)-limited ecosystems, yet the importance of this association in pristine boreal peatlands, which store 25 % of the world's soil (C), has been overlooked. External inputs of N to bogs are predominantly atmospheric, and given that regions of boreal Canada anchor some of the lowest rates found globally (*1 kg N ha -1 year -1), biomass production is thought to be limited primarily by N. Despite historically low N deposition, we show that boreal bogs have accumulated approximately 12-25 times more N than can be explained by atmospheric inputs.Here we demonstrate high rates of biological N 2 -fixation in prokaryotes associated with Sphagnum mosses that can fully account for the missing input of N needed to sustain high rates of C sequestration. Additionally, N amendment experiments in the field did not increase Sphagnum production, indicating that mosses are not limited by N. Lastly, by examining the composition and abundance of N 2 -fixing prokaryotes by quantifying gene expression of 16S rRNA and nitrogenase-encoding nifH, we show that rates of N 2 -fixation are driven by the substantial contribution from methanotrophs, and not from cyanobacteria. We conclude biological N 2 -fixation drives high sequestration of C in pristine peatlands, and may play an important role in moderating fluxes of methane, one of the most important greenhouse gases produced in peatlands. Understanding the mechanistic controls on biological N 2 -fixation is crucial for assessing the fate Responsible Editor: Matthew Wallenstein.Electronic supplementary material The online version of this article (doi:10.1007/s10533-014-0019-6) contains supplementary material, which is available to authorized users. Biogeochemistry (2014) 121:317-328 DOI 10.1007 of peatland carbon stocks under scenarios of climate change and enhanced anthropogenic N deposition.

Nitrogen (N) fixation by free-living bacteria is a primary N input pathway in many ecosystems and sustains global plant productivity. Uncertainty exists over the importance of N, phosphorus (P) and molybdenum (Mo) availability in controlling free-living N fixation rates. Here, we investigate the geographic occurrence and variability of nutrient constraints to free-living N fixation in the terrestrial biosphere. We compiled data from studies measuring free-living N fixation in response to N, P and Mo fertilizers. We used meta-analysis to quantitatively determine the extent to which N, P and Mo stimulate or suppress N fixation, and if environmental variables influence the degree of nutrient limitation of N fixation. Across our compiled dataset, free-living N fixation is suppressed by N fertilization and stimulated by Mo fertilization. Additionally, free-living N fixation is stimulated by P additions in tropical forests. These findings suggest that nutrient limitation is an intrinsic property of the biochemical demands of N fixation, constraining free-living N fixation in the terrestrial biosphere. These findings have implications for understanding the causes and consequences of N limitation in coupled nutrient cycles, as well as modeling and forecasting nutrient controls over carbon-climate feedbacks.

No abstract

1. Accurately quantifying rates and patterns of biological nitrogen fixation (BNF) in terrestrial ecosystems is essential to characterize ecological and biogeochemical interactions, identify mechanistic controls, improve BNF representation in conceptual and numerical modelling, and forecast nitrogen limitation constraints on future carbon (C) cycling.2. While many resources address the technical advantages and limitations of different methods for measuring BNF, less systematic consideration has been given to the broader decisions involved in planning studies, interpreting data, and extrapolating results. Here, we present a conceptual and practical road map to study design, study execution, data analysis and scaling, outlining key considerations at each step.3. We address issues including defining N-fixing niches of interest, identifying important sources of temporal and spatial heterogeneity, designing a sampling scheme (including method selection, measurement conditions, replication, and consideration of hotspots and hot moments), and approaches to analysing, scaling and reporting BNF. We also review the comparability of estimates derived using different approaches in the literature, and provide sample R code for simulating symbiotic BNF data frames and upscaling.4. Improving and standardizing study design at each of these stages will improve the accuracy and interpretability of data, define limits of extrapolation, and facilitate broader use of BNF data for downstream applications. We highlight aspects-such as quantifying scales of heterogeneity, statistical approaches for dealing with nonnormality, and consideration of rates versus ecological significance-that are ripe for further development. | 1123Methods in Ecology and Evoluঞon SOPER Et al.