Process-based and Data-based estimates of variable community compensation depth for ocean BGC model with special reference to Indian Ocean.

Ocean Carbon-Cycle Model Inter-comparison Project (OCMIP-II) provides accurate rendition of the annual mean carbon cycle for the global ocean. However it comes with a penalty of seasonal biases. Through this study we tried to capture the seasonality of carbon cycle in the model through process-based parameterization of community compensation depth (depth at which photosynthesis equals respiration of the whole biological community) and its retrieval via data-based inversion method using surface ocean pCO₂ and phosphate data. In the first method, by utilizing the Chl-a attenuated incoming solar radiation, a depth where solar radiation reaches 10 w m^-2 has been proposed as a method to obtain spatially and temporally varying Zc. The spatio-temporal varying Zc has improved the seasonality of the simulated CO₂ fluxes, surface ocean pCO₂, export and new production in the major upwelling zones of Indian Ocean. Analysis proved that better representation of biological exports and the modified nutrient profiles in the model supported the seasonal correction in the OCMIP -II protocol. This scheme captured the carbon cycle response to episodic upwelling with the related biological processes in the Indian Ocean.

In the second attempt the surface pCO₂ and phosphate observations has been utilized to infer the spatially and temporally varying Zc via a cyclo-stationary Bayesian inversion method. Indian Ocean has been divided into 8 bioprovinces with 12 months of seasonality for which a prior Zc of 75m is assumed. A cost function based on model and observation mismatch has been minimized by taking Zc as a control variable. The data-based and process-based estimates of variable Zc are consistent and retrieved a similar seasonal cycle for all bioprovinces with slight differences in amplitudes.

Major conclusions are:

(a) Seasonality in carbon cycle of OCMIP -II could be improved by varying Zc,

(b) A balance in model export and new production is required for a better seasonality of carbon cycle.

(c) Surface ocean pCO₂ observations can be used as better observational constraint for upwelling zones.

(d) Using surface observations we are able to retrieve inner biological model parameters.