Physical controls on spring bloom dynamics in the Irminger Basin, North Atlantic
Much of the primary production in northern latitudes is associated with the relatively short spring phytoplankton bloom. Quantifying the bloom is essential to understanding export production and energy transfer to higher trophic levels. This study focuses on the physical forcing controlling the spring bloom in the Irminger Basin (IB), situated between Greenland and Iceland. In situ data are available from four cruises to the region carried out under the UK Marine Productivity programme. This data set is extended with six years of SeaWiFS satellite chlorophyll-a concentration (chl-a) data, together with the corresponding model net heat flux (NCEP reanalysis) and satellite measured wind speed (QuikSCAT), sea surface temperature (SST; AVHRR) and photosynthetically available radiation (PAR; SeaWiFS). The remotely sensed data are complemented by a 1-D vertical mixing model and temperature and salinity profiles from Argo drifting profilers. The seasonality in temperature-nutrient (TN) relationships is investigated and the TN relationships are improved by including chlorophyll in the regressions. Basin-wide, daily estimates of nitrate, phosphate and silicate are made from satellite SST and chl-a. The study focuses on three biogeographical zones determined by cluster and Empirical Orthogonal Function analysis of SeaWiFS chl-a data. The three areas have distinct chl-a signatures and cover the East Greenland shelf, the Reykjanes Ridge and the Central Basin. An ANOVA analysis reveals that significant interannual variability is occurring in chlorophyll-a. An objective method for determining the start day of the spring bloom is described. Interannual variability in the timing of the initiation of the bloom and its magnitude and duration is discussed. The influence of the prevailing meteorology on chl-a in different seasons are investigated using generalized linear modelling. Whilst net heat flux and PAR are the dominating factors in spring, wind speed and SST become increasingly influential during summer and autumn. A method for estimating time series of Sverdrup’s critical depth from remotely sensed PAR and attenuation coefficient data is outlined. It is found that the spring bloom never begins before the mixed layer depth becomes shallower than the critical depth, and there is a delay of ~10 days. Specific criteria for the start of the bloom in terms of net heat flux and PAR are determined. The effect of nutrient depletion on the decline of the bloom is discussed. The East Greenland coastal zone is used as an example of the lasting impact that anomalous meteorological conditions can have on the following spring’s bloom. In 2002 the East Greenland region experienced anomalously low chl-a concentrations. Strong easterly winds, associated with the tip-jet phenomena, occurred throughout winter and spring and net heat flux was anomalously low in 2002. The spring bloom in the Irminger Basin can be affected by large scale climatic events, such as shifts in the North Atlantic Oscillation. Finally, the timing of nutrient depletion and its impact on community succession is considered. The possibility of iron limitation in the basin is discussed. A lower bound estimate of export production is made based on the timing of silica availability, and hence diatom dominance, of the spring bloom. The contributions to export production by diatoms and non-diatoms are estimated.