scholarly journals A derivative-free optimisation method for global ocean biogeochemical models

2021 ◽  
Author(s):  
Sophy Elizabeth Oliver ◽  
Coralia Cartis ◽  
Iris Kriest ◽  
Simon F. B. Tett ◽  
Samar Khatiwala
Keyword(s):  
Global Ocean ◽  
Biogeochemical Model ◽  
Biogeochemical Models ◽  
Derivative Free ◽  
Observational Uncertainty ◽  
Parameter Configuration ◽  
Ocean Biogeochemical Model ◽  
Observational Noise ◽  
Parameter Values ◽  
Earth System Models

Abstract. The performance of global ocean biogeochemical models, and the Earth System Models in which they are embedded, can be improved by systematic calibration of the parameter values against observations. However, such tuning is seldom undertaken as these models are computationally very expensive. Here we investigate the performance of DFO-LS, a local, derivative-free optimisation algorithm which has been designed for computationally expensive models with irregular model-data misfit landscapes typical of biogeochemical models. We use DFO-LS to calibrate six parameters of a relatively complex global ocean biogeochemical model (MOPS) against synthetic dissolved oxygen, inorganic phosphate and inorganic nitrate observations from a reference run of the same model with a known parameter configuration. The performance of DFO-LS is compared with that of CMA-ES, another derivative-free algorithm that was applied in a previous study to the same model in one of the first successful attempts at calibrating a global model of this complexity. We find that DFO-LS successfully recovers 5 of the 6 parameters in approximately 40 evaluations of the misfit function (each one requiring a 3000 year run of MOPS to equilibrium), while CMA-ES needs over 1200 evaluations. Moreover, DFO-LS reached a baseline misfit, defined by observational noise, in just 11–14 evaluations, whereas CMA-ES required approximately 340 evaluations. We also find that the performance of DFO-LS is not significantly affected by observational sparsity, however fewer parameters were successfully optimised in the presence of observational uncertainty. The results presented here suggest that DFO-LS is sufficiently inexpensive and robust to apply to the calibration of complex, global ocean biogeochemical models.

Using new observations and Machine Learning to improve organic sinking processes in the PlankTOM global ocean biogeochemical model 

2021 ◽  
Author(s):  
Anna Denvil-Sommer ◽  
Corinne Le Quéré ◽  
Erik Buitenhuis ◽  
Lionel Guidi ◽  
Jean-Olivier Irisson
Keyword(s):  
Organic Carbon ◽  
Carbon Cycle ◽  
Deep Ocean ◽  
Ecosystem Dynamics ◽  
Global Ocean ◽  
Biogeochemical Model ◽  
Carbon Export ◽  
Mixing Depth ◽  
Chl A ◽  
Biogeochemical Models

<p>A lot of effort has been put in the representation of surface ecosystem processes in global carbon cycle models, in particular through the grouping of organisms into Plankton Functional Types (PFTs) which have specific influences on the carbon cycle. In contrast, the transfer of ecosystem dynamics into carbon export to the deep ocean has received much less attention, so that changes in the representation of the PFTs do not necessarily translate into changes in sinking of particulate matter. Models constrain the air-sea CO<sub>2</sub> flux by drawing down carbon into the ocean interior. This export flux is five times as large as the CO<sub>2</sub> emitted to the atmosphere by human activities. When carbon is transported from the surface to intermediate and deep ocean, more CO<sub>2 </sub>can be absorbed at the surface. Therefore, even small variability in sinking organic carbon fluxes can have a large impact on air-sea CO<sub>2</sub> fluxes, and on the amount of CO<sub>2</sub> emissions that remain in the atmosphere.</p><p>In this work we focus on the representation of organic matter sinking in global biogeochemical models, using the PlankTOM model in its latest version representing 12 PFTs. We develop and test a methodology that will enable the systematic use of new observations to constrain sinking processes in the model. The approach is based on a Neural Network (NN) and is applied to the PlankTOM model output to test its ability to reconstruction small and large particulate organic carbon with a limited number of observations. We test the information content of geographical variables (location, depth, time of year), physical conditions (temperature, mixing depth, nutrients), and ecosystem information (CHL a, PFTs). These predictors are used in the NN to test their influence on the model-generation of organic particles and the robustness of the results. We show preliminary results using the NN approach with real plankton and particle size distribution observations from the Underwater Vision Profiler (UVP) and plankton diversity data from Tara Oceans expeditions and discuss limitations.</p>

scholarly journals A reactivity continuum of particulate organic matter in a global ocean biogeochemical model

2016 ◽  
Author(s):  
Olivier Aumont ◽  
Marco van Hulten ◽  
Matthieu Roy-Barman ◽  
Jean-Claude Dutay ◽  
Christian Ethé ◽  
...  
Keyword(s):  
Laboratory Experiments ◽  
Critical Role ◽  
Oxygen Demand ◽  
Sediment Oxygen Demand ◽  
Global Ocean ◽  
Biogeochemical Model ◽  
Mesopelagic Zone ◽  
Ocean Biogeochemical Model ◽  
The Impact ◽  
Reactivity Continuum

Abstract. The marine biological carbon pump is dominated by the vertical transfer of Particulate Organic Carbon (POC) from the surface ocean to its interior. The efficiency of this transfer plays an important role in controlling the amount of atmospheric carbon that is sequestered in the ocean. Furthermore, the abundance and composition of POC is critical for the removal of numerous trace elements by scavenging, a number of which such as iron are essential for the growth of marine organisms, including phytoplankton. Observations and laboratory experiments have shown that POC is composed of numerous organic compounds that can have very different reactivities. Yet, this variable reactivity of POC has never been extensively considered, especially in modeling studies. Here, we introduced in the global ocean biogeochemical model NEMO-PISCES a description of the variable composition of POC based on the theoretical Reactivity Continuum Model proposed by (Boudreau and Ruddick, 1991). Our model experiments show that accounting for a variable lability of POC increases POC concentrations in the ocean’s interior by one to two orders of magnitude. This increase is mainly the consequence of a better preservation of small particles that sink slowly from the surface. Comparison with observations is significantly improved both in abundance and in size distribution. Furthermore, the amount of carbon that reaches the sediments is increased by more than a factor of two, which is in better agreement with global estimates of the sediment oxygen demand. The impact on the major macro-nutrients (nitrate and phosphate) remains modest. However, iron (Fe) distribution is strongly altered, especially in the upper mesopelagic zone as a result of more intense scavenging: Vertical gradients in Fe are milder in the upper ocean which appears to be closer to observations. Thus, our study shows that the variable lability of POC can play a critical role in the marine biogeochemical cycles which advocates for more dedicated in situ and laboratory experiments.

scholarly journals Role of jellyfish in the plankton ecosystem revealed using a global ocean biogeochemical model

2021 ◽  
Vol 18 (4) ◽  
pp. 1291-1320
Author(s):  
Rebecca M. Wright ◽  
Corinne Le Quéré ◽  
Erik Buitenhuis ◽  
Sophie Pitois ◽  
Mark J. Gibbons
Keyword(s):  
Temporal Dynamics ◽  
Marine Ecosystem ◽  
Zooplankton Community ◽  
Ocean Model ◽  
Plankton Community ◽  
Global Ocean ◽  
Biogeochemical Model ◽  
Ocean Biogeochemical Model ◽  
Spatio Temporal ◽  
Plankton Community Structure

Abstract. Jellyfish are increasingly recognised as important components of the marine ecosystem, yet their specific role is poorly defined compared to that of other zooplankton groups. This paper presents the first global ocean biogeochemical model that includes an explicit representation of jellyfish and uses the model to gain insight into the influence of jellyfish on the plankton community. The Plankton Type Ocean Model (PlankTOM11) model groups organisms into plankton functional types (PFTs). The jellyfish PFT is parameterised here based on our synthesis of observations on jellyfish growth, grazing, respiration and mortality rates as functions of temperature and jellyfish biomass. The distribution of jellyfish is unique compared to that of other PFTs in the model. The jellyfish global biomass of 0.13 PgC is within the observational range and comparable to the biomass of other zooplankton and phytoplankton PFTs. The introduction of jellyfish in the model has a large direct influence on the crustacean macrozooplankton PFT and influences indirectly the rest of the plankton ecosystem through trophic cascades. The zooplankton community in PlankTOM11 is highly sensitive to the jellyfish mortality rate, with jellyfish increasingly dominating the zooplankton community as its mortality diminishes. Overall, the results suggest that jellyfish play an important role in regulating global marine plankton ecosystems across plankton community structure, spatio-temporal dynamics and biomass, which is a role that has been generally neglected so far.

Global ocean biogeochemical modelling with FESOM2-REcoM

2021 ◽  
Author(s):  
Ozgur Gurses ◽  
Judith Hauck ◽  
Moritz Zeising ◽  
Laurent Oziel
Keyword(s):  
Sea Ice ◽  
Finite Volume ◽  
Computational Efficiency ◽  
Ocean Model ◽  
Global Ocean ◽  
Biogeochemical Model ◽  
Coupled Models ◽  
Data Set ◽  
Biogeochemical Models ◽  
Mesh Resolution

<p>Marine biogeochemistry models are generally coupled to a physical ocean model. The biases in these coupled models can be attributed to simplified and empirical representation of biogeochemical processes, insufficient spatial mesh resolution which has an impact on the transport and mixing of biogeochemical substances in the ocean, and a deficit of physical parameterizations that intent to mimic unresolved processes such as eddies. Ocean Biogeochemical models based on variable mesh resolution proved to be convenient tools due to their computational efficiency and flexibility. Unlike standard structured-mesh ocean models, the mesh flexibility allows for a realistic representation of eddy dynamics in certain regions. Here, we present preliminary results of the coupling between the Finite-volumE Sea ice-Ocean Model (FESOM2.0) and the biogeochemical model REcoM2 (Regulated Ecosystem Model 2) in a coarse spatial resolution global configuration.<br>Surface maps of the simulated nutrients, chlorophyll a and net primary production (NPP) are comparable to available observational data sets. The control simulation forced with the JRA55-do data set reveals a realistic spatial distribution of nutrients, nanophytoplankton and diatom NPP, carbon stocks and fluxes. <br>FESOM2 utilizes a new dynamical core based on a finite-volume approach. The computational efficiency is about 2-3 times higher than the previous version FESOM1.4, whereas the quality of the simulated ocean and sea ice conditions and representation of biogeochemical variables are comparable in the two models. Thus, the new coupled model FESOM2- REcoM2 is very promising for ocean biogeochemical modelling applications.</p>

scholarly journals Swept under the carpet: organic matter burial decreases global ocean biogeochemical model sensitivity to remineralization length scale

2013 ◽  
Vol 10 (12) ◽  
pp. 8401-8422 ◽  
Author(s):  
I. Kriest ◽  
A. Oschlies
Keyword(s):  
Organic Matter ◽  
Numerical Models ◽  
Global Scale ◽  
Heterogeneous Data ◽  
Global Ocean ◽  
Biogeochemical Model ◽  
Length Scale ◽  
Benthic Fluxes ◽  
Biogeochemical Models ◽  
Wide Range

Abstract. Although of substantial importance for marine tracer distributions and eventually global carbon, oxygen, and nitrogen fluxes, the interaction between sinking and remineralization of organic matter, benthic fluxes and burial is not always represented consistently in global biogeochemical models. We here aim to investigate the relationships between these processes with a suite of global biogeochemical models, each simulated over millennia, and compared against observed distributions of pelagic tracers and benthic and pelagic fluxes. We concentrate on the representation of sediment–water interactions in common numerical models, and investigate their potential impact on simulated global sediment–water fluxes and nutrient and oxygen distributions. We find that model configurations with benthic burial simulate global oxygen well over a wide range of possible sinking flux parameterizations, making the model more robust with regard to uncertainties about the remineralization length scale. On a global scale, burial mostly affects oxygen in the meso- to bathypelagic zone. While all model types show an almost identical fit to observed pelagic particle flux, and the same sensitivity to particle sinking speed, comparison to observational estimates of benthic fluxes reveals a more complex pattern, but definite interpretation is not straightforward because of heterogeneous data distribution and methodology. Still, evaluating model results against observed pelagic and benthic fluxes of organic matter can complement model assessments based on more traditional tracers such as nutrients or oxygen. Based on a combined metric of dissolved tracers and biogeochemical fluxes, we here identify two model descriptions of burial as suitable candidates for further experiments and eventual model refinements.

scholarly journals Coastal-ocean uptake of anthropogenic carbon

2016 ◽  
Author(s):  
Timothée Bourgeois ◽  
James C. Orr ◽  
Laure Resplandy ◽  
Christian Ethé ◽  
Marion Gehlen ◽  
...  
Keyword(s):  
Coastal Ocean ◽  
Global Ocean ◽  
Biogeochemical Model ◽  
Temporal Data ◽  
Anthropogenic Carbon ◽  
Ocean Biogeochemical Model ◽  
Data Coverage ◽  
The Mean ◽  
Global Inventory ◽  
Offshore Transport

Abstract. Anthropogenic changes in atmosphere-ocean and atmosphere-land CO2 fluxes have been quantified extensively, but few studies have addressed the connection between land and ocean. In this transition zone, the coastal ocean, spatial and temporal data coverage is inadequate to assess its global budget. Thus we use a global ocean biogeochemical model to assess the coastal ocean's global inventory of anthropogenic CO2 and its spatial variability. We used an intermediate resolution, eddying version of the NEMO-PISCES model (ORCA05), varying from 20 to 50 km horizontally, i.e., coarse enough to allow multiple century-scale simulations but finer than coarse resolution models (~ 200 km), to begin to better resolve coastal bathymetry. Simulated results indicated that the global ocean absorbed 2.3 Pg C yr−1 of anthropogenic carbon during 1993–2012, consistent with previous estimates. Yet only 4.5 % of that (0.10 Pg C yr−1) is absorbed by the global coastal ocean, i.e., less than its 7.5 % proportion of the global ocean surface area. Coastal uptake is weakened due to a bottleneck in offshore transport, which is inadequate to reduce the mean anthropogenic carbon concentration of coastal waters to the mean level found in the open-ocean mixed layer.

scholarly journals Sensitivity of 21st-century projected ocean new production changes to idealized biogeochemical model structure

2021 ◽  
Author(s):  
Genevieve Jay Brett ◽  
Daniel B. Whitt ◽  
Matthew C. Long ◽  
Frank Bryan ◽  
Kate Feloy ◽  
...  
Keyword(s):  
Nutrient Uptake ◽  
21St Century ◽  
Nutrient Utilization ◽  
Global Ocean ◽  
Biogeochemical Model ◽  
New Production ◽  
Biogeochemical Models ◽  
Projected Changes ◽  
Organic Particles ◽  
Future Work

Abstract. While there is agreement that global warming over the 21st century is likely to influence the biological pump, Earth system models (ESM) display significant divergence in their projections of future new production. This paper quantifies and interprets the sensitivity of projected changes in new production in an idealized global ocean-biogeochemistry model. The model includes two tracers that explicitly represent nutrient transport, light- and nutrient-limited nutrient uptake by the ecosystem (new production), and export via sinking organic particles. Globally, new production declines with warming due to reduced surface nutrient availability, as expected. However, the magnitude, seasonality, and underlying dynamics of the nutrient uptake are sensitive to the light and nutrient dependencies of uptake, which we summarize in terms of a single biological timescale that is a linear combination of the partial derivatives of production with respect to light and nutrients. Although the relationships are non-linear, this biological timescale is correlated with several measures of biogeochemical function: shorter timescales are associated with greater global annual new production and higher nutrient utilization. Shorter timescales are also associated with greater declines in global new production in a warmer climate and greater sensitivity to changes in nutrient than light. Future work is needed to characterize more complex ocean biogeochemical models in terms of similar timescale generalities to examine their climate change implications.

Sign in / Sign up

Export Citation Format

Share Document