Aquatic Sciences Meeting, Albuquerque 2001

PC08 New Techniques and Technologies from Single Cells to the Global Ocean

Date: Wednesday, February 14, 2001

Location: Southwest Hall

Bernard, O, , INRIA, France, France, obernard@sophia.inria.fr

Sciandra, A, , CNRS, Villefranche-sur-mer, France, sciandra@obs-vlfr.fr

Software sensor as a new tool to monitor the internal nitrogen quota of algae

Software sensors, issued from the non linear automated control field and initially developed for non-linear chemical systems, have been applied to a continuous culture of phytoplankton. A software sensor combines an analytical differential equation based model and partial measurements of the system, in order to estimate the evolution of the non-measured variables. It filters data and estimates the actual state of living systems. The efficiency of this approach is illustrated with a nitrate-limited chemostat experiment with the chlorophyceae Dunaliella tertiolecta performed in a computer controlled fluctuating environment. The Droop model has been used to represent the algal dynamics, and the on-line biomass estimation preformed by an automated HIAC cell counter feeds the software sensor. It provides estimates for the remaining nitrate concentration and for the internal cell quota. The software predictions match quite well the direct measurements even when the cells are supplied with a slow varying fluctuating nitrogen source. For faster time scales, the Droop model seems not to be appropriate to serve as a basis for the state estimation. These results open new insights for the in situ estimation of the algal status.

This Session Listing

Aquatic Sciences Meeting, Albuquerque 2001

Quicklinks