Abstract
Abstract. Uncertainties in projections of marine biogeochemistry from Earth system models (ESMs) are associated to a large degree with the imperfect representation of the marine plankton ecosystem, in particular the physiology of primary and secondary producers. Here, we describe the implementation of an optimality-based plankton–ecosystem model (OPEM) version 1.1 with variable carbon : nitrogen : phosphorus (C:N:P) stoichiometry in the University of Victoria ESM (UVic; Eby et al., 2009; Weaver et al., 2001) and the behaviour of two calibrated reference configurations, which differ in the assumed temperature dependence of diazotrophs. Predicted tracer distributions of oxygen and dissolved inorganic nutrients are similar to those of an earlier fixed-stoichiometry formulation in UVic (Nickelsen et al., 2015). Compared to the classic fixed-stoichiometry UVic model, OPEM is closer to recent satellite-based estimates of net community production (NCP), despite overestimating net primary production (NPP), can better reproduce deep-ocean gradients in the NO3-:PO43- ratio and partially explains observed patterns of particulate C:N:P in the surface ocean. Allowing diazotrophs to grow (but not necessarily fix N2) at similar temperatures as other phytoplankton results in a better representation of surface Chl and NPP in the Arctic and Antarctic oceans. Deficiencies of our calibrated OPEM configurations may serve as a magnifying glass for shortcomings in global biogeochemical models and hence guide future model development. The overestimation of NPP at low latitudes indicates the need for improved representations of temperature effects on biotic processes, as well as phytoplankton community composition, which may be represented by locally varying parameters based on suitable trade-offs. The similarity in the overestimation of NPP and surface autotrophic particulate organic carbon (POC) could indicate deficiencies in the representation of top-down control or nutrient supply to the surface ocean. Discrepancies between observed and predicted vertical gradients in particulate C:N:P ratios suggest the need to include preferential P remineralisation, which could also benefit the representation of N2 fixation. While OPEM yields a much improved distribution of surface N* (NO3--16⋅PO43-+2.9 mmol m−3), it still fails to reproduce observed N* in the Arctic, possibly related to a misrepresentation of the phytoplankton community there and the lack of benthic denitrification in the model. Coexisting ordinary and diazotrophic phytoplankton can exert strong control on N* in our simulations, which questions the interpretation of N* as reflecting the balance of N2 fixation and denitrification.
Highlights
Earth system models (ESMs) are routinely used for simulating both the possible future development and the past of our climate system (e.g. IPCC, 2013; Hülse et al, 2017; Keller et al, 2018; Park et al, 2019)
The above description of the model behaviour highlights some of the improvements of our optimality-based (OPEM, optimality-based plankton– ecosystem model (OPEM)-H) compared to the original biogeochemistry in the University of Victoria ESM (UVic) model
Some of these may be possible with the original UVic with improved parameters, e.g. the deep-ocean N : P distribution (Fig. 8) or a better global net community production (NCP) (Fig. 12), as these vary strongly among our different parameter sets tested during the calibration process of OPEM and OPEM-H (Chien et al, 2020)
Summary
Earth system models (ESMs) are routinely used for simulating both the possible future development and the past of our climate system (e.g. IPCC, 2013; Hülse et al, 2017; Keller et al, 2018; Park et al, 2019). The optimality concept is based on the “assumption that natural selection should tend to produce organisms optimally adapted to their environments” (Smith et al, 2011), which is applicable to marine plankton, where intense mixing and the absence of physical boundaries ensure strong competition, and short generation times allow for rapid evolution These formulations have shown their ability to describe ecosystem behaviour in 0-D and 1-D modelling studies (e.g. FernándezCastro et al, 2016; Su et al, 2018), and to predict patterns of phytoplankton nutrient and light co-limitation based on satellite and in situ observations (Arteaga et al, 2014).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
