Abstract
Abstract. The biogeochemical evaluation toolkit, BGC-val, is a model- and grid-independent Python toolkit that has been built to evaluate marine biogeochemical models using a simple interface. Here, we present the ideas that motivated the development of the BGC-val software framework, introduce the code structure, and show some applications of the toolkit using model results from the Fifth Climate Model Intercomparison Project (CMIP5). A brief outline of how to access and install the repository is presented in Appendix A, but the specific details on how to use the toolkit are kept in the code repository. The key ideas that directed the toolkit design were model and grid independence, front-loading analysis functions and regional masking, interruptibility, and ease of use. We present each of these goals, why they were important, and what we did to address them. We also present an outline of the code structure of the toolkit illustrated with example plots produced by the toolkit. After describing BGC-val, we use the toolkit to investigate the performance of the marine physical and biogeochemical quantities of the CMIP5 models and highlight some predictions about the future state of the marine ecosystem under a business-as-usual CO2 concentration scenario (RCP8.5).
Highlights
It is widely known that climate change is expected to have a significant impact on weather patterns, the cryosphere, the land surface, and the ocean (Stocker et al, 2015; Cook et al, 2013; Le Quéré et al, 2013; Rhein et al, 2013)
Due to the high thermal capacity of water, most of the excess heat captured by the greenhouse effect is absorbed by the ocean (Rhein et al, 2013). This increases the temperature of the waters, which causes sea levels to rise via thermal expansion (Church et al, 2013), may accelerate the melting of sea ice (Moore et al, 2015), and may push many marine organisms outside of their thermal tolerance range (Poloczanska et al, 2016)
For each Earth system model submitted to CMIP5, the development team chose how they wanted to divide the ocean into a grid composed of individual cells
Summary
It is widely known that climate change is expected to have a significant impact on weather patterns, the cryosphere, the land surface, and the ocean (Stocker et al, 2015; Cook et al, 2013; Le Quéré et al, 2013; Rhein et al, 2013). BGC-val has been used to compare various ocean models against a wide range of marine datasets, including the Takahashi air–sea flux of CO2 (Takahashi et al, 2009), the European Space Agency Climate Change Initiative (ESA-CCI) ocean colour dataset (Grant et al, 2017), the World Ocean Atlas data for temperature (Locarnini et al, 2013), salinity (Zweng et al, 2013), oxygen (Garcia et al, 2013a), and nutrients (Garcia et al, 2013b), and the MAREDAT (Buitenhuis et al, 2013b) global database for marine pigment (Peloquin et al, 2013), picophytoplankton (Buitenhuis et al, 2012), diatoms (Leblanc et al, 2012), and mesozooplankton (Moriarty and O’Brien, 2013) These datasets are all publicly available and are typically distributed as a monthly climatology or annual mean NetCDF file
Sign-in/Register to view highlights & summary 
Published Version (
Free)
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 call