During LITFASS98 airborne measurements from the helicopter-borne turbulence probe Helipod and from the research aircraft Dornier 128 were taken to determine area-averaged vertical turbulent fluxes. For LITFASS98 a square-shaped relatively small flight pattern of 10 km x 10 km for the Helipod and 15 km x 15 km for the Dornier aircraft was flown at three altitudes above heterogenous ground. The lowest flight path was at 140 m altitude for the Helipod and at 245 m for the Dornier aircraft. The main goal was to determine vertical latent and sensible heat fluxes through inverse modeling using only measurements form the lowest flight path. The inverse method uses appropriate model parameters that are based on theoretical assumptions to fit measured data. The output of the inverse model provides the gradient and the temporal development of the mean potential temperature and of the humidity. Inserting the gradients into the budget equation and using linear extrapolation to the ground values for the turbulent fluxes can be obtained. In a second step turbulent fluxes from inverse modeling were compared to turbulent fluxes retrieved with an ordinary method based on measurements on three heights and on ground measurements. It appears that using the inverse modeling method has the advantage of not needing the approximation of a linear extrapolation for the complete boundary layer as it is necessary for retrieving turbulent fluxes from airborne measurements in three heights. Also the inverse modeling method does not require further measurements (e.g. ground or mast measurements). Simply one square in low altitude has to be flown. In future the approach to use inverse modeling will be extended and improved on further airborne measurements. Future aims include the possibility to be able to obtain fluxes of momentum as well as horizontal fluxes of latent and sensible heat.