Diversity in the Simulation of Chlorophyll Concentration by CMIP5 Earth System Models over the Indian Ocean

Abstract

Among many other physical and biogeochemical processes, oceanic upwelling is a major contributor to the primary productivity of oceans. In the present study, sixteen models from the ‘Coupled Climate Model Inter-comparison Project phase 5’ (CMIP5) are assessed for their capability in simulating the Chlorophyll (Chl-a) concentration against satellite observations over the northern Indian Ocean. The sixteen CMIP5 models are categorized into three groups based on their relative skill, Group-A models had the highest skill and captured the phase of the bloom during summer monsoon season whereas the Group-B and Group-C models mostly failed to reproduce the Chl-a concentrations. The observed interannual variations were poorly simulated by all the CMIP5 models. Group-A models showed a negative bias in Chl-a concentration over the northern Arabian Sea and a positive bias in Chl-a simulation off Somalia over the western Indian Ocean. High Chl-a associated with the coastal upwelling along the west coasts of India and Sri Lanka was poorly simulated by CMIP5 models. The study highlights the regional deficiency in CMIP5 climate models in simulating Chl-a and the need for improved coupled physical-biogeochemical models over the tropical Indian Ocean.

KEYWORDS:

• Chlorophyll
• CMIP5 models
• Indian Ocean
• primary productivity
• upwelling

Acknowledgements

The authors acknowledge the Ocean Colour Climate Change Initiative (OC-CCI) dataset used in this study made available by the European Space Agency through the link http://www.esa-oceancolour-cci.org/. The Coupled Model Intercomparison Project 5 (CMIP5) is produced and made freely available under the World Climate Research Programme (WCRP). The High Performance Computing (HPC) facility provided by IIT Delhi and supported by Department of Science and Technology (DST-FIST, 2014), Govt. of India are thankfully acknowledged. Graphics generated in this manuscript using PyFerret and R.

Disclosure statement

Authors declare no conflict of interest.

Authors’ contributions

VS and VP conceptualized the study. VS downloaded and plotted the data. VS and VP analyzed model results. VS wrote the first draft and VP made numerous corrections and finalized the manuscript with the help of co-author.