An algorithm to determine primary production in Lake Constance by remote sensing

Abstract

Any model for aquatic primary production requires the phytoplankton biomass as most important input parameter. However, only the near-surface chlorophyll concentration can be derived from remote sensing data, whereas the vertical distribution of phytoplankton biomass has to be modelled according to the optical properties of the respective water body. Further input parameters that cannot be derived by remote sensing are the parameters of the photosynthesis-light-curve of the involved phytoplankton. Data on the photosynthetic parameters are rare, since measurements are costly and time consuming. However, at Lake Constance a long-term data set exists of primary production measured by the radiocarbon method, and other relevant parameters. This is an ideal data base to develop an algorithm for primary production, since the following issues can be adressed beforehand by data analysis: 1) How can vertical profiles of phytoplankton biomass and production be derived from near-surface chlorophyll considering water temperature and the vertical attenuation coefficient, and 2) what impact does each of the photosynthetic parameters have on the depth integral of photosynthesis? The light saturation parameter Pmax was fairly well predictable by considering the actual water temperature. By contrast, the dependency of the spectrally sensitive parameter alpha on changes in the spectral composition of the incident light could not be satisfactorily described by the existing data set. However, alpha has a major contribution to the depth integral of photosynthesis. Therefore, new measurements of the photosynthetic parameters will be performed by PAM fluorometry to validate the algorithm.