The development of satellite derived nitrate and stratification indices for the southern Benguela ecosystem

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

An earth observation based study was conducted in the southern Benguela upwelling system, aimed at developing remotely sensed proxies to determine ecological conditions conducive to the formation of harmful algal blooms (HAB) which are endemic to the system. The aim of this study was to identify the relationship between nutrient availability, turbulence and phytoplankton community assemblages using remotely sensed data. Certain phytoplankton functional groups are adapted to a particular environment within an ecological space conceptually based on the nutrient availability and turbulence (Margalef, 1978). Two proxies, representing the nutrient availability, and turbulence or stratification, were created using satellite-derived surface nitrate (NO3) concentrations and the 12ºC isotherm (Iso12) depths, and used to define the ecosystem state through Margalef's Mandala (Margalef, 1978). The approach involved the development of robust algorithms using in situ data collected in the greater St Helena Bay region to estimate the surface NO3 concentrations and the depth of the Iso12 for the southern Benguela, using remotely sensed sea surface temperatures (SST) and wind data. The derivation of the nutrient proxy was based on a model initially developed by Dugdale et al., (1989) then modified by Silió-Calzada et al., (2008) for use in the Benguela. The turbulence proxy was derived using a simple linear regression model to estimate the depth of the Iso12 which was utilized as a proxy for the thermocline depth in the system. The performance of the nutrient and turbulence proxies were assessed on local, meso- and synoptic scales for their ability to resolve the event and seasonal scale variations in the inner shelf environment of the southern Benguela. The derived NO3 and Iso12 products were sufficiently able to resolve the event and mesoscale variability of the system in 2005, 2006 and 2007. The performance of the products at capturing the annual and intra-seasonal variability of the system was satisfactory, displaying an ability to resolve the ecosystem upwelling variability. Using the NO3 and Iso12 products as the nutrient and turbulence proxies was satisfactory as a first attempt at using earth observation to classify the ecosystem according to Margalef's Mandala (Margalef, 1978). The proxies were able to model the ecosystem inner-shelf environment for multiple years and thus create the ability to hypothesize ecosystem sub-habitats occupied by particular life forms of phytoplankton. There were however two concerns that needed further consideration in the approach: 1).The large warm bias discovered between the in situ and remotely sensed temperatures which had a direct influence on the validity of the algorithms in the ecosystem and 2.) The temporal and spatial disconnects between the physical forcing and biological response of the ecosystem and subsequent impact upon the utility of the remotely sensed proxies.