Seasonal variability of the oceanic upper layer and its modulation of biological cycles in the Canary Island region

The Canary Island region is rich in mesoscale phenomena that affect cycles of physical and biological processes. A 1D version of the Regional Oceanic Modeling System (ROMS) is used south of the Gran Canaria Island to simulate seasonal climatologies of these cycles. The model is forced with monthly air–sea fluxes averaged from 1993 to 2002 and initialized with mean in situ profiles of temperature, salinity, oxygen and nitrate concentrations. The K-Profile Parameterization (KPP) mixed layer submodel is compared with other submodels using idealized numerical experiments. When forced with realistic air–sea fluxes, the model correctly reproduces the annual cycle of temperature (mixed layer depth), with minimum surface values of 18 °C (maximal depth > 105 m) in February during convective mixing resulting from a negative heat flux. Maximum temperatures above 23 °C (minimal depth < 20 m) are simulated from September to October after strong summer heating and a decrease in Trade Winds intensity. A simple ecosystem model is coupled to the physical model, which provides simulated biological cycles that are in agreement with regional observations. A phytoplankton bloom develops in late winter, driven by the injection of new nutrients into the euphotic layer. Simulated chlorophyll shows a deep maximum fluctuating around 100 m with concentrations around 1 mg Chla /m³, while surface values are low (around 0.1 mg Chla /m³) during most of the year. The physical and biological model results are validated by comparisons with data from regional studies, climatological fields and time-series from the ESTOC station.