A Nutrients-Phytoplankton-Zooplankton-Detritus (NPZD) ecosystem model is coupled to the Miami Isopycnic-Coordinate Ocean Model (MICOM) to study plankton dynamics in the Arabian Sea.Experiments oriented to testing the NPZD-MICOM coupled model sensitivity to variations in several parameters are performed. Particular attention is paid to the rates of detritus sinking and maximum phytoplankton growth rate. The coarse features of the blooms are captured by all the experiments considered, and agree with earlier models for the biological activity of the region. Intensity, duration, and peaks of the blooms are found to be quite sensitive to the parameter choices. The existence of an offshore deep chlorophyll maximum in the model is found to be closely related to the increase of the detritus sinking rate. The use of temperaturedependent maximum growth rate of phytoplankton and the increase of detritus sinking rate are shown to improve model results compared with in situ and satellite observations.The differences between the present results and those from previous modelling efforts in the region, where few-or multi-layer hydrodynamical models or even fixed-level hydrodynamical models have been employed, are found to be in the same range as the differences among the results from the various sensitivity experiments presented here. This indicates that a small uncertainty in the knowledge of the ecosystem model parameters can be more important in ecosystem modelling than the uncertainty associated to the differences in the vertical coordinate representation of the underlying hydrodynamical model.There are still biological processes, such as denitrification which affect the vertical distribution of nutrient concentrations, that are not included neither in the previous works in the area nor in the present work. This might mask any potential advantages of the present MICOM-NPZD coupled model, especially at mid-depths in the ocean.