Temporal Variability of Trichodesmium spp. and Diatom-Diazotroph Assemblages in the North Pacific Subtropical Gyre

White, Angelicque E.; Watkins-Brandt, Katie S.; Church, Matthew J.

doi:10.3389/fmars.2018.00027

Cited by 37 publications

(73 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of variations in nifH gene copy number per cell in different diazotrophs and methodological issues (e.g., nonspecificity of primers and probes used in qPCR), conversion to cell numbers is not straightforward. More details can be found in Text S1 in the supporting information (see also White et al, 2018). available radiation (PAR), mixed layer depth (MLD), averaged PAR in the mixed layer (PAR mld ), chlorophyll-a concentration ([Chl]), and modeled surface iron concentration (Fe).…”

Section: Updating the Global Diazotrophs Databasementioning

confidence: 99%

Data‐Driven Modeling of the Distribution of Diazotrophs in the Global Ocean

Tang

Cassar

2019

Geophysical Research Letters

View full text Add to dashboard Cite

Diazotrophs play a critical role in the biogeochemical cycling of nitrogen, carbon, and other elements in the global ocean. Despite their well‐recognized role, the diversity, abundance, and distribution of diazotrophs in the world's ocean remain poorly characterized largely due to limited observations. Here we update the database of diazotroph nifH gene abundances and assess how environmental factors may regulate diazotrophs at the global scale. Our meta‐analysis more than doubles the number of observations in the previous database. Using linear and nonlinear regressions, we find that the abundances of Trichodesmium, UCYN‐A, UCYN‐B, and Richelia relate differently to temperature, light, and nutrients. We further apply a random forest algorithm to estimate the global distributions of these diazotrophic groups, identifying undersampled potential hot spots of diazotrophy in the South Atlantic and southern Indian Ocean, and in coastal waters. The distinct ecophysiologies of diazotrophs highlighted here argue for separate parameterizations of different diazotrophs in model simulations.

show abstract

Section: Updating the Global Diazotrophs Databasementioning

confidence: 99%

Data‐Driven Modeling of the Distribution of Diazotrophs in the Global Ocean

Tang

Cassar

2019

Geophysical Research Letters

View full text Add to dashboard Cite

show abstract

“…One trait that is widely variable in nature is filament length. It has been widely accepted that the average filament length is 100 cells (46) but more recent studies have suggested that they are typically much shorter, with a geometric mean of 13.2 ± 2.3 cells per filament, but with a mean range of 1.2 to 685 cells per filament in situ (47). Conditions for in situ sampling are widely variable so we hypothesized that filament length plays a role in maintaining growth in different environments: low light, low CO 2 and low N 2 .…”

Section: Resultsmentioning

confidence: 99%

“…One of the many advantages of using this multi-paradigm framework is that we can simulate emergent behavior of a population. In situ data reports a wide mean range of trichome length from 1.2 to 685 cells (47); we used the model to investigate possible causes because this is a difficult phenotype to investigate experimentally. Our simulations suggest that even though longer filaments suffer from diffusional effects that limit growth, they are better able to handle stress (Figure 6) consistent with literature.…”

Section: Discussionmentioning

confidence: 99%

MultIscale MultiObjective Systems Analysis (MIMOSA): an advanced metabolic modeling framework for complex systems

Gardner

Boyle

2019

Preprint

View full text Add to dashboard Cite

12In natural environments, cells live in complex communities and experience a high degree of 13 heterogeneity internally and in the environment. Unfortunately, most of the metabolic modeling 14 approaches that are currently used assume ideal conditions and that each cell is identical, limiting 15 their application to pure cultures in well-mixed vessels. Here we describe our development of 16MultIscale MultiObjective Systems Analysis (MIMOSA), a metabolic modeling approach that can 17 track individual cells in both space and time, track the diffusion of nutrients and light and the 18 interaction of cells with each other and the environment. As a proof-of concept study, we used 19 MIMOSA to model the growth of Trichodesmium erythraeum, a filamentous diazotrophic 20 cyanobacterium which has cells with two distinct metabolic modes. The use of MIMOSA 21 significantly improves our ability to predictively model metabolic changes and phenotype in more 22 complex cell cultures. 23 linear optimization problems to satisfy species and community level objectives, leveraging the 47 inner solution as a constraint for the outer problem. However, this approach still relies on a priori 48 determination of relative objective preference as well as predetermination of both species-level 49 and community-level objectives. Additionally, cells are treated as homogenous spatial groups (13) 50 or homogenous species groups (14), which limits the accurate simulation of cells acting 51 individually, interacting with their environment, and ultimately forming communities. These 52 approaches thus discount the complexity of individual cells forming communities and, instead of 53 acting uniformly with neighbors or species, create dynamic intercellular and inter-environmental 54 reactions (13,(18)(19)(20). 55 56To more accurately model the complexity of community growth, a new modeling approach must 57 be developed. We have developed MultIscale MultiObjective Systems Analysis (MIMOSA), an 58 advanced metabolic modeling framework for complex systems. This approach uses a multi-scale 59 multi-paradigm metabolic modeling approach can leverage simple, powerful stoichiometric 60 metabolic models and integrate spatio-temporal tracking of cells, nutrient diffusion, cell-cell 61 interactions and cell-environmental interactions. This approach requires the use of both continuous 62 and discrete variables as well as several different mathematical formalisms to reflect the multilevel 63 behavior in populations. Therefore, we use an agent-based modeling (ABM) framework to allow 64 direct interaction of different levels through the encapsulation of physiological, environmental, 65 and metabolic models. ABM is a bottom-up modeling approach; the model is made up of a set of 66 agents, which are allowed to act independently as long as they follow distinct rules of behavior 67 defined by the user, this allows us to simulate emergent behavior of complex communities that 68 arise from individual agent behaviors (21-24). The system behavior emerges as a result of th...

show abstract

“…This source water was pre‐filtered through 200 μm Nitex mesh in order to exclude grazers. This mesh also excludes large colonies of nitrogen fixers, but, at least for Trichodesmium , this effect is expected to be small (∼ 6% of the total; White et al ). Water was then fed through a plankton net with 20 μm mesh net and cod‐end.…”

Section: Methodsmentioning

confidence: 99%

Nitrogen fixation rates diagnosed from diurnal changes in elemental stoichiometry

Follett

White

Wilson

et al. 2018

Limnology & Oceanography

Self Cite

View full text Add to dashboard Cite

The carbon, nitrogen, and phosphorus (C, N, and P, respectively) composition and elemental ratios were measured in the 20–200 μm size fraction during July 2015 in the surface waters of an anticyclonic eddy encountered north of Hawaii in the oligotrophic North Pacific Subtropical Gyre. The observed particulate N : P ratio fluctuated by approximately a factor of two over the diel cycle. We present a simple mathematical argument connecting this change to a rate of biological nitrogen fixation, and calculate the nitrogen fixation rate to be ≥ 13 nmol L−1 d−1 for this size class. This value is higher than simultaneous bottle‐incubation based rates measured with isotopic tracers, yet is consistent with historic rate measurements from the region. As confirmation of our methods, diurnal changes in C : N : P of laboratory cultures of the diazotrophic genus Trichodesmium were measured. In the laboratory, we show that estimates of nitrogen fixation from stoichiometric time series are equivalent to those derived directly from mass balance. The disparity between nitrogen fixation rates derived from tracer measurements and particulate stoichiometry in the field suggests that large diazotrophs may be underestimated in small volume (∼ 4 L) bottle incubations as a result of either spatial heterogeneity or vertical migration of large cells. Otherwise, processes other than diazotrophy must cause the observed changes in stoichiometry. This approach represents a novel and scalable means of quantifying in situ nitrogen fixation rates from diurnal changes in size‐fractionated stoichiometry. We also infer carbon fixation, growth rates, and phosphorus uptake in the 20–200 μm size class.

show abstract

Temporal Variability of Trichodesmium spp. and Diatom-Diazotroph Assemblages in the North Pacific Subtropical Gyre

Cited by 37 publications

References 53 publications

Data‐Driven Modeling of the Distribution of Diazotrophs in the Global Ocean

Data‐Driven Modeling of the Distribution of Diazotrophs in the Global Ocean

MultIscale MultiObjective Systems Analysis (MIMOSA): an advanced metabolic modeling framework for complex systems

Nitrogen fixation rates diagnosed from diurnal changes in elemental stoichiometry

Contact Info

Product

Resources

About