Quoted from: Totterdell, Ian J. "Description and evaluation of the Diat-HadOCC model v1. 0: the ocean biogeochemical component of HadGEM2-ES." Geoscientific Model Development 12, no. 10 (2019): 4497-4549. https://doi.org/10.5194/gmd-12-4497-2019 

The number of general circulation models (GCMs) available to study climate has increased rapidly in recent years, and the range of processes and feedbacks that they can represent has also become more comprehensive. Initially there were just physical models describing the circulation of the atmosphere and the ocean and how those circulations redistributed and stored heat, as well as the response of the system to rising atmospheric CO2. The first coupled climate model to include representations of the land and marine carbon cycles, including terrestrial vegetation and soils and marine ecosystems, capable of representing their basic feedbacks on the climate was HadCM3LC (Cox et al., 2000). In that model, the terrestrial vegetation was described by the TRIFFID model (Cox, 2001), while the chemistry of carbon dioxide in seawater and the marine ecosystem was described by the Hadley Centre Ocean Carbon Cycle (HadOCC) model (Palmer and Totterdell, 2001). The latter is a simple nutrient–phytoplankton–zooplankton–detritus (NPZD) model using nitrogen as the limiting element.

A brief overview of Met Office model nomenclature is useful here. The Met Office modelling system used (over a time period of several decades) for climate studies and for numerical weather prediction is known as the Unified Model, and the coupled climate models exist as various versions of it. The HadCM3LC model mentioned above featured a lower-resolution (“L”) ocean sub-model than the HadCM3C model, which itself was the member of the HadCM3 family of coupled climate models (Gordon et al., 2000; version 4.5 of the Unified Model) that featured an interactive carbon cycle (“C”) in the atmosphere, on land and in the ocean. The HadGEM2 family of climate models (The HadGEM2 Development Team, 2011), a development of HadCM3 with enhanced resolution and improved parameterisations that was used for CMIP5 simulations, was version 6.6 of the Unified Model. In particular, HadGEM2-ES (Collins et al., 2011), featuring active Earth system components including version 1.0 of the Diat-HadOCC sub-model, was version 6.6.3.

The aim of this paper is to describe and validate version 1.0 of the Diat-HadOCC model, as used in HadGEM2-ES to run simulations for the CMIP5 experiment. Although the simulations were run several years ago this description of the model is important as a record and can inform other modellers of potential parameterisations that succeeded (or not) here. The equations are presented and described in detail, and reasons are given for certain choices made in the representation of processes and in the values of parameters. Where potential other uses of the model (e.g. in ocean-only simulations forced by reanalysis fluxes) differ from its use here, this is mentioned. The publicly available model output submitted to CMIP5 is used to evaluate the model, and its successes and weaknesses are discussed.