The use of spaceborne medium resolution imaging spectrometers with neural network algorithms has proven a large potential for application with optically complex inland waters. We make use of this approach to investigate the bio-physical dynamics in a eutrophic lake, applying three different neural networks to a dataset of 16 images acquired in June through August 2011. Concurrent in-situ data are measured by means of automatically deployed instruments from a moored platform, resolving the vertical distribution of various parameters at sub-daily temporal resolution. Phytoplankton blooms occur in different stratification layers, allowing the assessment of their influence on remote sensing estimates. A qualitative synopsis of the biophysical processes in the lake is given, but parameterization with in-situ attenuation profiles and accurate IOP estimates is needed to significantly enhance quantitative matchup comparisons. Recommendations on the combination of in-situ and satellite measurements are therefore given as an outlook.