Research papersAn improved coastal upwelling index from sea surface temperature using satellite-based approach – The case of the Canary Current upwelling system
Introduction
Eastern Boundary Upwelling Ecosystems (EBUEs) extend over temperate and tropical regions and include some of the most productive ecosystems in the world (Carr and Kearns, 2003). Oceanography of EBUEs is forced by the equatorward trade winds that lead to the upward pumping of cold intermediate water. The phenomenon can distinctively be identified by seasonally variable low sea surface temperature (SST) and nutrient-rich water in coastal area, compared to the average SST at the same latitude (Wooster et al., 1976).
The high primary productivity of coastal upwelling systems (Herbland and Voituriez, 1974, Minas et al., 1982) sustains large fisheries of small pelagic species, a major economic resource that contributes to 20% of the global fish production for less than 3% of the world oceans surface (Ryther, 1969, Cushing, 1971, Fréon et al., 2009a).
The Canary Current Upwelling System, that extends from the Iberian Peninsula (43°N) to the south of Senegal (8°N, Fig. 1), is one of the world׳s four major EBUEs (Barton, 1998, Arístegui et al., 2009). The seasonal variability of the trade winds between winter and summer (Fig. 2a) induces pronounced coastal SST anomalies (Fig. 2b) with a high seasonal variability, mainly in the southern part of the system (Fig. 2b), first described by Wooster et al. (1976) based on historical measurements, wind stress and currents from on-board observations.
As described by Ekman (1905) and later by Bakun (1973), the cool and rich upwelled water along the coastal shelf edge is a consequence of the alongshore trade wind forcing. The strength and position of the wind depends on the latitudinal migration of the Intertropical Convergence Zone (ITCZ) and the associated Azores High pressure area, both oscillating between their northermost and southernmost positions in summer/winter respectively (Fig. 1), generating a seasonal wind and SST pattern (Wooster et al., 1976, Mittelstaedt, 1991, Nykjær and Van Camp, 1994, Van Camp et al., 1991).
The wind vectors from the CCMP (Cross-Calibration Multi-platform) dataset (Fig. 2a) shows a variable wind circulation between winter and summer. The coastal region is characterized by upwelling favorable wind stress all year-round (Bakun and Nelson, 1991), with maximum values during summer north to 21°N and during winter south to 21°N (Mittelstaedt, 1991).
In most of the study area (10–30°N), the near-surface wind is under influence of northeast trade wind, as part of the lower part of the Hadley circulation cell. North of 30°N the exact position of a zonal belt of high pressure cells is predominantly determined by the location of the Azores High and is characterized by low north-easterly surface winds (Benazzouz, 2014). South of 15°N the ITCZ – the meridional convergence of the northeast and southeast trade winds – becomes the main driver of the upwelling seasonality.
In the Iberian Peninsula (37–43°N), equatorward winds dominate from early June to late September and generate an equatorward mean upwelling surface flow associated to the Portugal surface current (Mittelstaedt, 1991).
The physical -and classic- way of quantifying the upwelling intensity is by computing the Coriolis forces that induces the Cross-Shore Ekman transport (CSET). The wind-based “Coastal upwelling index” can be interpreted as the water flux theoretically transported offshore by the wind stress from the coastal upward flux of colder water, assuming an infinite ocean (Ekman, 1905). Consequently, this upwelling index do not consider bathymetricy of the continental shelf and therefore cannot render the complexity of the two dimensional spatial structure of the upwelling such as coastal upwelling cells, cape effects and filaments.
Another limitation in the computation of wind-based upwelling indices is the relatively short length (<20 years) of time series available, excepting of composite products such as the CCMP Ocean surface winds, available from July 1987.
Satellite-derived SST provides a quantitative and synoptic overview of thermal features in the ocean, including the wind-driven coastal SST. The surface cooling from the upwelled water reaching the sea surface is therefore a potential proxy of the upwelling intensity.
Surface wind flux and coastal SST are two complementary parameters to quantify the spatial extent and intensity of the upwelling process at synoptic scale, the first one from a physical theory and the second one based from direct observation. The mixing of the cold upwelled water flux with the warmer surrounding surface water by turbulent mixing should be taken into account as well as the understanding of the causal mechanism of the SST variability from the wind flux. Because of their intrinsic links, Ekman transport and the resulting SST anomaly can be used as complementary variables to investigate the status of the upwelling.
The aim of the paper is to develop a new methodology to compute an improved SST-based upwelling index, applicable for all EBUEs. A review of various methods based on satellite and wind to quantify upwelling dynamics is conducted. Moreover, some of past studies concerned both the spatial structure of upwelling (e.g., fronts, eddies, filaments) and its fluctuations on synoptic to interannual time-scales (e.g., Van Camp et al., 1991; Nykjær and Van Camp, 1994; Kostianoy and Zatsepin, 1996, Demarcq and Faure, 2000, Santos et al., 2005, Nieto et al., 2012) in less than 20-year time series. Several satellite SST data series in EBUEs are available at a suitable spatio-temporal resolution (<10 km and 1 month) and in a continuous coverage and synoptic monitoring. We used a 30 year (1981–2011) time series of 8-day global SST with the aim of synthesizing more spatially detailed information than the past works.
Section snippets
Datasets
We use two homogeneous data sets from spatial observation: a 30 years sea surface temperature (SST) series from the AVHRR (Advanced Very High Resolution Radiometer) and two wind data series, respectively the SeaWind sensor and the global wind CCMP product which combined from all sensors available from July 1987 until today.
Methods
To investigate the average seasonal cycle of the SST, an 8-day climatological year is constructed from the 8-day data series from September 1981 to December 2011.
Two similar 8-day surface wind climatologies are constructed from both QuikSCAT and CCMP winds products by averaging daily images respectively from July 1999 to October 2009 and January 1988 to December 2011.
Seasonal pattern of SSTmin and SSTmax
The seasonal variability of the observed SSTmin (Fig. 8a) strongly varies latitudinally because it depends first on the seasonality of the heat exchanges with the atmosphere, and second on the intensity of the upwelling itself, that also varies seasonally.
Because they result from the interplay between these two parallel processes, the absolute values of SSTmin have little signification by themselves and only their seasonality is described here.
In the same line, it has been noted that the
Spatial considerations
The spatial variability of the coastal upwelling structure is explored, with a special focus on the effects of the bathymetry. The Fig. 3 shows typical spatial patterns of the upwelled waters, for the Moroccan region (left panel) and the Senegalese–Mauritanian region (right panel), respectively during summer (August) and late winter (March) when the upwelling reach its respective maximum latitudinal extensions.
It has been shown (Fig. 5a) that there is little variability in the spatial location
Conclusion
The proposed SST-based upwelling index is not only a satisfying alternative to the Cross-shore Ekman transport, it supplies a complementary description of the upwelling physical properties, in particular independent measurements of its intensity and thermal inertia, that varies in space and time with intensity and duration. Detailed patterns of the variability of the Canary upwelling are reproduced, that integrates useful informations on the complexity of the effects of both bathymetry and
Acknowledgments
We applied the FLAE approach for the sequence of authors. We thank the 50th Anniversary Young African fellowship programme of IOC (Intergovernmental Oceanographic Commission) as well as the French Institute of Research for Development (IRD) for partially supporting this work. The SST data were provided by GHRSST and the US National Oceanographic Data Center, in a project partly supported by a grant from the NOAA Climate Data Record (CDR). QuikSCAT and CCMP global wind were obtained from the
References (51)
- et al.
Sub-regional ecosystem variability in the Canary Current upwelling
Prog. Oceanogr.
(2009) - et al.
Temporal variability observed in the hydrographic regime near Cabo coreiro in the nortwest Africa upwelling region Februry to April 1974
Deep Sea Res.
(1977) - et al.
Production regimes in four Eastern Boundary Current systems
Deep-Sea Res. II
(2003) - et al.
Comparison between CCMP, QuikSCAT and buoy winds along the Iberian Peninsula coast
Remote Sens. Environ.
(2013) Upwelling and the production of fish
Adv. Mar. Biol.
(1971)- et al.
Eastern boundary upwelling ecosystems: integrative and comparative approaches
Prog. Oceanogr.
(2009) - et al.
The West African coastal upwelling filaments and cross-frontal water exchange conditioned by them
J. Mar. Syst.
(1996) - et al.
L’upwelling de la côte atlantique du Maroc entre 1994 et 1998
Comptes Rendus Geosci.
(2005) The upwelling area off Africa – a description of phenomena related to coastal upwelling
Prog. Oceanogr.
(1983)The ocean boundary along the northwest African coast: circulation and oceanographic properties at the sea surface
Prog. Oceanogr.
(1991)