Remote sensing and modelling in upwelling systems

Abstract

Optical instruments on board of satellites can measure phytoplankton biomass in the ocean for primary productivity studies. But these measurements may lead to erroneous results if the vertical structure of phytoplankton in the water column-a deep maximum in the algal concentration-is not taken into account.
We present a model of the distribution and growth of phytoplankton in the water column for the upwelling system off-western Iberia intended to properly interpret and complement remote sensing observations of algal biomass.
Winds data from the ERS-1 scatterometer and ECMWF are used to asses upwelling conditions together with AVHRR abservations of the sea surface temperature and to estimate the surface mixed layer depth and mixing in the upper water column.
Keywords: upwelling,phytoplankton,optics,deep chlorophyll maximum, scatterometer

Introduction

The western Iberian coast is characterized by a wind climate with two regimes: a mainly westerly-southerly one during winter, and one with an equatorward component during the rest of the year. The northerly wind reaches its maximum strength and steadiness during Summer and, as a result of this regime, wind driven upwelling conditions occur from March through October.
Coastal upwelling brings cold, nutrient-rich waters from the subsurface layers to the surface which are advected off-shore by the mean, wind-induced circulation. Primary productivity is enhanced as a result of this regime and the optical properties of these waters are modified by the pigments present in phytoplanktonic cells, mainly chlorophyll-a.
Optical sensors on board of satellites can measure these modifications providing estimations of pigment content which constitute indirect measures of algal biomass.

Satellite derived pigment (mg/m3).
This image of pigment concentration during strong upwelling conditions in western Iberia (14/9/80) was obtained by the Coastal Zone Colour Scanner, which was operative between 1978-1986. The thermal image corresponding to this date shows similar structures.

The pigment concentration retrieved by means of remote ocean colour sensors is a value representative of the upper metres of the water column (the first attenuation depth), but vertical profiles of pigment concentration usually show a non homogeneous vertical structure. A deep chlorophyll maximum is a common feature which very often can be partially or completely undetected by such instruments. In upwelling systems the maximum deepens as the waters are advected offshore and depleted from nutrients in the upper part of the water column.
To attain a proper interpretation of remote ocean colour observations we developed a model of the distribution and growth of phytoplankton in the water column as a result of the particular conditions present in an upwelling regime.
Satellite measurements of sea-surface temperature (SST) are used to force such a model as they contain information about upwelling conditions, off-shore advection and warming of upwelled waters, <A/and can be used to map nutrient distributions.
Upwelling parameters and mixed layer conditions are also obtained from wind fields derived from the ERS-1 Scatterometer and ECMWF models.

Satellite derived SST (Centig.).
This satellite image of SST is a composite of data from the Advanced Very High Resolution Radiometer obtained between 9/9/80 and 14/9/80. Note that the cold water structures in this image are similar to the high pigment structures in the 'colour' image of 14/9/80 above.

In Situ Observations off Western Iberia

Data from a number of cruises off Western Iberia are used for model developement.As an example we present results from the MORENA 3 campaign, 26/7/94 - 11/8/94, during upwelling favourable winds.

AVHRR DERIVED SST, 28/07/94

Satellite derived SST obtained on 28 July 1994. Cold waters originate in the upwelling center near 41.5 N. The stations sampled on 27 and 28 July during the MORENA 3 cruise are indicated together with the 180 m isobath. Results from these observations are shown below.

Zonal profiles at 41 N and 42 N reveal the upward lifting of isotherms and high nutrients towards the coast associated with upwelling. Note the progressive deepening of the chlorophyll maximum away from the near-shore source of upwelled waters. Both the lack of nutrients in the upper water column and sinking of phytoplankton contribute to this phenomena.

TEMPERATURE, NITRATE AND CHLOROPHYLL ALONG 41 N, 27-28/07/94

Contours of Temperature, Nitrate concentration and Chlorophyll concentration at 41.2 N obtained from the stations sampled on 26-27 July 1994. z coordinate in the vertical, East longitude in the horizontal axis. Similar results were obtained from the stations sampled on 6-8 August,1994.

Modelling the growth and distribution of phytoplankton

The goal of this model is to simulate the vertical distribution and growth of phytoplankton under the intermittent input of new nutrients and turbulence due to coastal upwelling and the subsequent depletion of nutrients and stratification of the water column as it is advected offshore and warmed up by solar heating. A simple 1-D formulation is assumed to account for the mainly vertical variability within any parcel of water which will be associated with a pixel as observed from a satellite. Only one nutrient, Nitrogen, is considered at this stage. A simplified cell-quota theory which allows 'luxury' uptake of nutrients and internal nutrient-controlled growth is also considered in view of the non-steady conditions of upwelling regimes. Such a formulation may help to simulate the observed persistence of upwelling structures in ocean color images when their SST signature has vanished in thermal infrared images due to solar heating. Primary productivity in the water column can thus be calculated as the rate of increase of biomass and per-surface or per-day values obtained by integration.
Phytoplankton biomass in terms of chlorophyll concentration Chl, dissolved nutrient concentration N and phytoplankton internal nutrient concentration PON as functions of z and t are described by the equations (1), (2) and (3) below, respectively:

Turbulent mixing in the upper water-column depends on the structure of the upper layer, from which we estimate the vertical profile of K. The surface boundary layer in upwelling regimes usually consists of 1)a surface mixed layer whose depth is scaled as

and depends strongly on the wind, 2) a transition zone below the mixed layer where the flow is near-critical and mixing still occurs and 3)an interior where the flow is stable.

The model reproduces fairly well the dynamic response of phytoplankton to the vertical advection of new nutrients. The development and vertical migration of a peak of chlorophyll following the optimum conditions of light and nutrients and the magnitude of the modelled biomass at the peak compare well with observations off-Western Iberia.

Two simulations are shown below.
In the first one the history of vertical profiles (depth in the vertical axis) of chlorophyll (green), dissolved Nitrogen (red) and Phytoplankton Nitrogen (black) is tracked for ten days. Active upwelling close in-shore is simulated for the first 48 hs with initial conditions derived from in-situ observations.

SIMULATION OF AN UPWELLING-RELAXATION EVENT

SIMULATION OF AN UPWELLING-RELAXATION EVENT

Model Simulation: An upwelling event starts at t=0 hrs. When the water parcel leaves the zone of active upwelling (plots 1,2 and 3) the peak in chlorophyll moves down due to sinking and nutrient limitation in the upper part of the water column.

The second simulation shows the spatial distribution of chlorophyll in a vertical plane along a zonal transect at 41.5 N. Initial conditions were derived from the MORENA 3 observations and the AVHRR derived SST for 28/7/94.

MODELLED CHLOROPHYLL FOR THE AVHRR SST FIELD 28/07/94

Modelled Chlorophyll along a transect at 41.5 N. SST from the AVHRR image 28/7/94 was used to assess upwelling conditions. A time proportional to the observed warming since upwelling ocurred was derived for each pixel and used to run the model. Data from the MORENA-3 cruise provided initial conditions for nutrient and chlorophyll distributions in the zone of active upwelling. The sloping bathymetry towards the coast is in black.

Conclusions

The main features of phytoplankton dynamics in the Western Iberian upwelling system are reproduced with this approach. The model provides a means to interpret the near-surface remote estimations of phytoplankton biomass as the upper part of a dynamic vertical distribution. The approach requires the synergistic use of satellite estimations of sea surface temperature and wind fields. The future generation of ocean colour sensors, together with the existing ERS and NOAA instruments, will provide a suitable tool to asess the rol of upwelling systems in the global carbon cycle.

Acknowledgments

HRPT AVHRR data from the NOAA satellites were provided by the University of Dundee and processed into Sea Surface Temperature maps at the Satellite Oceanography Laboratory of the University of Hawaii and at the Oceanography and Coastal Management Laboratory of the Universidad Nacional, Costa Rica.
Data from the MORENA cruises were provided after analysis by Dr. F.F. Perez, Dr. X.A. Alvarez-Salgado, Dr. C. Castro and Dr. M.X. Fernandez, from the Instituto de Investigaciones Marinas from Vigo, Spain.
Wind fields from the ERS-1 Scatterometer were provided by the Department of Oceanography from Space (LOS+CERSAT) at IFREMER, France.
Wind fields from the ECMWF models were provided by the British Atmospheric Data Centre

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry