The photosynthetic carbon fixation by phytoplankton – the micro algae populating the upper lit layers of the ocean – fuels the oceanic food web and affects the global ocean carbon cycle and CO₂ levels. Assessing the causes of the variability of phytoplankton concentrations is, therefore, crucial to predict their future evolution in response to warming of the World Ocean. Understand the coupling between physics and biology is essential to properly quantify the marine-ecosystem variability.

The objective of the present project is to provide new elements to understand the response of phytoplankton to climate decadal changes through its physical forcing to help forecasting phytoplankton evolution in future. At interannual to decadal scales, an inverse relationship between chlorophyll concentration (Chl; a proxy for phytoplankton biomass) and sea surface temperature (SST; an indicator of stratification) has been reported over more than 60% of the global ocean using remote sensing. This inverse relationship is usually expected in nutrient-limited subtropical regions, because a warming-induced stratification would reduce the upward nutrient supply and then productivity. However, in the Northeast Atlantic a parallel increase of Chl and SST from the beginning of the 1980s to the 2000s has been reported, associated with a regime shift of the Atlantic Multidecadal Oscillation from a cold to a warm phase. This parallel increase suggests that SST may not be always the best indicator of stratification in this region.

Therefore, in the project it has been investigated how seasonal cycles of phytoplankton have changed in the Northeast Atlantic 40o-0oW and 30o-50oN) using the satellite reprocessed archive from [Antoine et al., 2005], who applied the same algorithms and an adapted calibration to the CZCS (1979-1983) and SeaWiFS (1998-2002) satellite ocean color missions, providing a homogenous time series. The variability of the mixed-layer depth (MLD) derived from in situ vertical temperature profiles since 1941, as well as wind stress data (a dynamical driver of the MLD) extracted from an extensive collection of surface marine data since 1800 (ICOADS) were also investigated.

Changes in the seasonal cycles of Chl and MLD between 1979-1983 and 1998-2002 appear over the most productive region of the study (40o-0oW and 40o-50oN). In this area, the Chl seasonal cycle was characterized in 1979-1983 by a fall bloom slightly stronger than the spring bloom. Over 1998-2002, the fall bloom was weaker and the spring bloom was dominant compared to the CZCS period. This dominant spring bloom is preceded by a maximum of MLD (MLD_max) that has deepened between the two time periods.

The spring bloom starts earlier at lower latitudes (30o-40oN) where winter mixing is less vigorous and the ecosystem is likely to be limited by nutrients rather than by light. Further north (40o-50oN), where the ecosystem is likely to be limited by light, Chl_max occurs later, when the MLD rises above a critical depth. During the SeaWiFS time period, the amplitude of the spring bloom was stronger and high chlorophyll values extended further south. This southward extension followed an increase (up to 80 m) of the MLD_max values reached in winter between 1979-1983 and 1998-2002. This deepening of MLD_max can be related to an increase of the wintertime winds between the two time periods and would have allowed a more efficient uplift of nutrient. The fall bloom, which is barely reported in literature, clearly appeared during the CZCS era. Twenty years later, it has strongly decreased, high Chl values were restricted to the northern boundary, Chl_max was reached 2-3 months earlier while the MLD deepening occurred one month later at the end of summer. Combined with a deeper MLD_max, this delayed deepening was likely to increase light limitation and limit the full development of the fall bloom.

The decadal increase of SST and Chl related with the unexpected (at first sight) deepening of MLD reported above highlights the complexity of the links between SST changes, stratification, and biological responses. The MLD and wind changes that have been observed on two 5-year time periods separated by 15 years, are in agreement with similar results reported over a longer time period (1960-2004) in this region. An interesting prospect would be to determine if phytoplankton changes observed here are also consistent over a longer time period. Regarding the Phytoplankton Color Index (an index of Chl issued from the Continuous Plankton Record survey) combined with the ESA Globcolour satellite product of Chl, it appears that Chl have increased since the 1980s in the Northeast Atlantic. The next step of the project is to focus on this long-term tendency regarding the physical forcing derived from in situ and satellite data.