Physical forcing and phytoplankton phenology over NOW polynya from 1998 to 2013, as observed from ocean color data Christian Marchese1, Camille Albouy1, Steve Vissault1, Dany Dumont2, Fabrizio D’Ortenzio3, Jean-Éric Tremblay4, Simon Bélanger1 1Université

du Québec à Rimouski, 2Institut des sciences de la mer de Rimouski, 3Laboratoire d'océanographie de Villefranche, 4Université Laval

Study Area

Methodology

Main conclusions

Within polar regions, polynyas are recognized as particular and unique areas. The NOW polynya (Fig.1) located in northern Baffin Bay, between Ellesmere Island and Greenland, is one of the largest polynyas in the Northern Hemisphere and the most biologically productive. Its formation is mainly due to the consolidation of an ice bridge at 79°N. The NOW is closely connected to Lancaster Sound (where a smaller polynya is also present) and Baffin Bay by a system of ocean currents (Fig. 2) that directly influence the region’s climate and biology.

•  To characterize phytoplankton phenology (e.g. start time, duration and amplitude), four Gaussian models were employed on a pixel-by-pixel basis. The most complex model used was a two-peak Gaussian with the addition of a linear term that define the rate of decrease (increase) in chlorophyll-a concentration over time (Zhai et al. 2011).

•  Due to different local environmental conditions that are present in the study area, regional differences in timing and duration of the pelagic phytoplankton are found.

•  The models were fitted to the chlorophyll-a time series using the Levenberg-Marquardt algorithm (LMA) for non-linear regression. •  The AICc (AIC with a correction for finite sample sizes) was used for model selection. •  In the study area, the bloom start was calculated by averaging the results obtained from three different methods (see Navarro et al. 2012): - through the simple calculation of ±2 the standard deviation - when the chlorophyll-a concentration reaches 20% of the amplitude - when the chlorophyll-a concentration rises 5% above the annual median values

•  The bloom start timing appear to be correlated with the ice melt onset and the stabilization of the upper mixed layer. •  The length of open water period together with the increase in wind storms may modulate the bloom duration.

Results Bloom start climatology (1998-2013)

Bloom duration vs. Bloom start

Julian day Fig. 1 (© PEW - Oceans North Canada)

Bloom duration vs. Open water period

Fig. 2 (© PEW - Oceans North Canada)

Aims Summarizing the possible impact of sea ice phenology and wind speed on the pelagic phytoplankton bloom timing and duration.

Number of stormy days vs. time periods

Bloom duration vs. Number of stormy days

Bloom start vs. ice-retreat timing

Satellite data Merged (SeaWiFS/MODIS/MERIS) Chlorophyll-a from ESA GlobColour projects (globcolour.info) Sa-ice concentration data from NSIDC (nsidc.org) Wind speed data from IFREMER (cersat.ifremer.fr)

Time periods •  •  •  • 

1998 – 1999 2000 – 2001 2002 – 2003 2004 – 2005

References •  •  •  • 

2006 – 2007 2008 – 2009 2010 – 2011 2011 – 2012

Zhai, L., Tang, C.L.C., Platt, T., Sathyendranath, S., Walls, R.H. (2011). Phytoplankton phenology on the Scotian Shelf. ICES Journal of Marine Science, 68, 781-791. Navarro, G., Caballero, I., Prieto, L., Vázquez, A., Flecha, S., Huertas, I.E., Ruiz, J. (2012). Seasonal-to-interannual variability of chlorophyll-a bloom timing associated with physical forcing in the Gulf of Cádiz. Adv. Space Res., 50, 1164-1172.