**1. Introduction**

122 Remote Sensing of Biomass – Principles and Applications

Sathyendranath, S., Watts, L., & Devred, E. (2004). Discrimination of diatoms from other phytoplankton using ocean-colour data. *Mar. Ecol. Prog. Ser.* 272: 59–68 Schoeld, O., Bergmann, T., Grzymski, J. & Glenn, S. (1999). Spectral uorescence and

Sellner, S.G., Sellner, K.G., & Brownlee, E.F. (2003). Effects of Barley Straw (*Hordeum vulgare*)

Smayda, T.J. & Reynolds, C.S. (2003). Strategies of marine dinoflagellate survival and some

Stumpf, R.P., Culver, M.E., Tester, P.A., Tomlinson, M., Kirkpatrick, G.J. & B.A. Pederson*.*

Stumpf, R.P. & Tomlinson, M.C. (2005). Remote sensing of harmful algal blooms. In Remote

Subramanian, A., Carpenter, E.J., Karentz, P.G. & Falkowski, PG. (1999a). Optical properties

Totterdell, I.J., Armstrong, R.A., & Drange, H. (1993).Trophic resolution. In Evans, G. T. and

Uitz, J., Claustre, H., Morel, A., Hooker, & S.B. (2006). Vertical distribution of phytoplankton

Vidussi, F., Claustre, H., Manca, B.B., Luchetta, A., & Marty, J.C. (2001). Phytoplankton

Mediterranean Sea during winter. *J Geophys Res* 106 (C9): 19,939–19,956 Vila, M., Giacobbe, M.G., MasÓ M., Gangemi E., Penna A., Sampedro N., Azzaro F., Camp

minutum in two Mediterranean coastal areas. *Harmful algae* 4: 673-695 Worm, B., Sandow, M., & Oschlies, A. (2005). Global patterns of predator diversity in the

Part II: Topical Studies in Oceanography 47 (12-13): 2363-240

Wright, S.W. & RL van den Enden. (2000). Phytoplankton community structure and stocks

Wright, S.W. & Jeffrey, S.W. (2006). Pigment markers for phytoplankton production.

spectral photosynthetic characteristics. *Limnol. Oceanogr.* 44: 608-617 Subramanian, A., Carpenter, E.J. & Falkowski, P.G. (1999b) Optical properties of the marine

NATO ASI, Vol. I 10. Springer-Verlag, Berlin, pp. 71–92

McKee, pp. 277–96. AH Dordrecht, The Netherlands: Springer. 347 pp. Subramanian, A., & Carpenter, E.J. (1994). An empirically derived protocol for the detection

Sieburth, J. MCN. 1979. *Sea microbes.* New York: Oxford University Press 491 pp.

color imagery and other data. *Harmful Algae News* 35: 1–14

*SPIE Ocean Optics* XIV 3:60–67

*Remote Sensing* 158: 1559–1569

*Geophys. Res*. 111:C08005

sensing. Limnol. Oceanogr. 44: 618-627

open oceans. *Science* 309: 1365–1369

*Handbook Env Chem* 2N: 71-104

rules of assembly. *J Sea Res* 49: 95–106

*Phycology* 15: 525-531

inherent optical properties during upwelling events off the coast of New Jersey.

on Freshwater and Brackish Phytoplankton and Cyanobacteria. *J Applied* 

(2003). Monitoring *Karenia brevis* blooms in the Gulf of Mexico using satellite ocean

Sensing of Coastal Aquatic Environments, ed. RL Miller, CE Del Castillo, BA

of blooms of the marine cyanobacterium *Trichodesmium*using CZCS imagery. *Int. J.* 

of the marine diazotrophic cyanobacteria *Trichodesmium* spp. I. Absorption and

diazotrophic cyanobacteria *Trichodesmium* spp. II. Reflectance model for remote

Fasham, M. J. R. (eds), Towards a Model of Ocean Biogeochemical Processes.

communities in open ocean: An assessment based on surface chlorophyll. *J.* 

pigment distribution in relation to upper thermocline circulation in the eastern

J., & Galluzzi L. (2005). A comparative study on recurrent blooms of Alexandrium

in the East Antarctic marginal ice zone (BROKE survey, January-March 1996) determined by CHEMTAX analysis of HPLC pigment signatures. *Deep Sea Research*

The ocean is a highly variable system affected by a large range of processes that spans the continuous spectra of spatial and temporal scales (Wunsch, 1996). The spatial scales of variation range from basin-wide gyres (thousands of kilometers) to turbulence (less than a meter), and the time scales from those that are climate related (decades) to short-term processes (seconds). Information on this whole range of processes is required for the comprehension of the marine system dynamics. Despite the continuous advances in technology, remote sensing is the only observing platform capable of providing continuous information on biological and physical properties over vast areas of the ocean. With some limitations, the regular and repeated coverage offered by satellites is still unachievable through in situ measurements.

Because the ocean is largely opaque over much of the usable electromagnetic spectrum, the ability of satellites to capture ocean properties is generally confined to the surface. Nevertheless, satellite-borne sensors provide us with a relatively large range of measurements such as sea surface color, sea surface height, sea surface temperature, sea surface winds, sea surface salinity, waves, and to a lesser extent, current fields. The availability, for the first time, of time series expanding for several years or decades at regional and global ocean scales has changed our perception of the ocean (Barber & Hilting, 2000). A majority of these measurements is restricted to physical properties such as temperature, sea level or sea surface roughness and inferred variables (currents, winds, etc.). The only routinely acquired satellite measurement providing information on ocean biological processes is sea surface color. Since early measurements obtained by the Coastal Zone Color Scanner (CZCS), sea color sensors have provided quantitative information on the distribution of surface chlorophyll (CHL) concentration (an index of phytoplankton biomass) at regional to global scales and its variability in space and time (e.g., Abbott & Zion, 1987; Antoine et al., 2005; Behrenfeld et al., 2001). This information is relevant to estimate the ocean productivity, a key factor for understanding the dynamics of pelagic foodwebs and some aspects of climate change.

Using SVD Analysis of Combined Altimetry and Ocean Color

coupling in the Mediterranean Sea.

to the understanding of larger areas.

**2. Data and methodology 2.1 Altimetry satellite data** 

gray lines.

Satellite Data for Assessing Basin Scale Physical-Biological Coupling in the Mediterranean Sea 125

formation and water mass hydrological properties, the frontal currents and mesoscale eddies, and the topographic and coastal influence, add complexity to the physical-biological

Fig. 1. Bathymetry of the Mediterranean Sea. The 200 and 2000 m isobaths are shown with

In this work, we analyze the basin scale patterns of phytoplankton variability at interannual, seasonal and intra-annual scales and the associated driving forces in the Mediterranean Sea based on 12 years of concurrent ocean color and altimetry satellites data. The knowledge of the phytoplankton variability and its relation to ocean circulation is critical to understand marine ecosystem dynamics and biogeochemical cycles, with implications ranging from marine food webs to climate change. The physical mechanisms that regulate phytoplankton patterns in the Mediterranean are analogous to those in larger oceanic areas and therefore comprehension of the processes occurring therein are pertinent

The satellite-borne altimetry is initially designed to estimate the sea surface height (SSH) by measuring the satellite-to-surface round-trip time of a radar pulse. These measurements however include the Earth's geoid which varies by tens of meters across the ocean and is not accurately estimated (Fu et al., 1994). This unknown geoid is removed from satellite observations by subtracting a long term mean of the altimeter measurements from the observations. However, this procedure removes also the mean dynamic SSH and satellite measurements refer only to SLA, which contains information on the geostrophic current fields, mesoscale eddy variability and changes in the thermocline depth (Le Traon et al., 1998; Stammer, 1997). Uncertainties on the location of the satellite on its orbit and disturbances of the radar pulse by the atmosphere introduce additional errors in the SLA

The merging of ocean color datasets with other satellite measurements providing information of ocean dynamics has potential benefits to the understanding of some aspects of the coupling between fluid driven processes and plankton dynamics. In this regard, altimetry satellite data allows the characterization of sea level anomaly (SLA), containing information on the geostrophic current fields, mesoscale eddy variability and changes in the thermocline depth (Le Traon et al., 1998; Stammer, 1997). Lehahn et al. (2007) and Leterme & Pingree (2008) investigated the effect of the geostrophic velocity derived from satellite SLA on the redistribution of satellite CHL. Also, the impact of mesoscale eddies on the spatial patterns of CHL has been analyzed by Siegel et al. (1999, 2008).

Further insight into the basin-scale dynamics affecting CHL and SLA, as an indicator of changes in the thermocline depth, in the equatorial Pacific was provided by Wilson & Adamec (2001). Correlations between CHL and SLA data using empirical orthogonal function (EOF) analysis show different responses associated with El Niño-Southern Oscillation phases. In the global ocean, direct correlations between CHL and SLA are predominately negative as lower SLA implies thermocline weakening and consequent mixing of the water column, which results in increased nutrient flux to the surface layer and phytoplankton biomass enhancement (Wilson & Adamec, 2002). However, there are areas in all ocean basins where positive correlations suggest that CHL is affected by processes other than thermocline variations. For example, Uz et al. (2001) found positive correlations between satellite CHL and SLA associated with the propagation of Rossby waves. These studies exemplify how multiple satellite observations are used to understand basin-scale dynamics and their impacts on the ocean phytoplankton biomass variability. Following these works, Jordi et al. (2009) used the singular value decomposition (SVD) analysis of the cross-covariance matrix between satellite CHL and SLA to analyze the regional scale dynamics in the northwestern Mediterranean Sea. Their results highlight the role of the water mass transported by the regional circulation on the variability of the phytoplankton biomass. The SVD analysis may be superior to EOF analysis in identifying correlated spatial patterns between pairs of spatial time series (Bretherton et al., 1992).

In the Mediterranean Sea, a semi-enclosed marginal sea with limited geographical dimensions (Fig. 1), ocean color data reveals that oligotrophic conditions prevail for most of the year (D'Ortenzio et al., 2002). Biological production is mainly regulated by physical processes enhancing nutrient supply to surface layers and by allochtonous inputs from the continents and the atmosphere (Barale et al., 2008). Satellite ocean color data in the Mediterranean Sea has demonstrated dominance of the seasonal cycle in phytoplankton biomass (Bosc et al., 2004). With some regional variations, the typical temperate-latitude cycle with maximum biomass in late winter-spring and minimum during summer stratified conditions occurs throughout most of the basin (Bricaud et al., 2002). Inter-annual variations in CHL concentrations are also noticeable both at a local scale and over the whole basin, and have been related to climatic fluctuations (D'Ortenzio et al., 2003).

Complementarily, altimetry satellite data shows that the sea level variability in the Mediterranean is a complex combination of a wide range of spatial and temporal scales (Cazenave et al., 2001; Larnicol et al., 2002). Besides the marked seasonal cycle in SLA caused by the steric effect, important intra- and inter-annual signals are observed associated with permanent or transitory oceanographic structures such as frontal currents and mesoscale eddies (Jordi & Wang, 2009; Pujol & Larnicol, 2005). The multiple driving forces including the ocean–atmosphere interaction, the phenomenology of the deep water formation and water mass hydrological properties, the frontal currents and mesoscale eddies, and the topographic and coastal influence, add complexity to the physical-biological coupling in the Mediterranean Sea.

Fig. 1. Bathymetry of the Mediterranean Sea. The 200 and 2000 m isobaths are shown with gray lines.

In this work, we analyze the basin scale patterns of phytoplankton variability at interannual, seasonal and intra-annual scales and the associated driving forces in the Mediterranean Sea based on 12 years of concurrent ocean color and altimetry satellites data. The knowledge of the phytoplankton variability and its relation to ocean circulation is critical to understand marine ecosystem dynamics and biogeochemical cycles, with implications ranging from marine food webs to climate change. The physical mechanisms that regulate phytoplankton patterns in the Mediterranean are analogous to those in larger oceanic areas and therefore comprehension of the processes occurring therein are pertinent to the understanding of larger areas.
