Atmospheric Correction Performance of Hyperspectral Airborne Imagery over a Small Eutrophic Lake under Changing Cloud Cover

摘要

Atmospheric correction of remotely sensed imagery of inland water bodies is essentialudto interpret water-leaving radiance signals and for the accurate retrieval of water quality variables.udAtmospheric correction is particularly challenging over inhomogeneous water bodies surrounded byudcomparatively bright land surface. We present results of AisaFENIX airborne hyperspectral imageryudcollected over a small inland water body under changing cloud cover, presenting challenging butudcommon conditions for atmospheric correction. This is the first evaluation of the performance ofudthe FENIX sensor over water bodies. ATCOR4, which is not specifically designed for atmosphericudcorrection over water and does not make any assumptions on water type, was used to obtainudatmospherically corrected reflectance values, which were compared to in situ water-leavingudreflectance collected at six stations. Three different atmospheric correction strategies in ATCOR4udwas tested. The strategy using fully image-derived and spatially varying atmospheric parametersudproduced a reflectance accuracy of �0.002, i.e., a difference of less than 15% compared to the in situudreference reflectance. Amplitude and shape of the remotely sensed reflectance spectra were in generaludaccordance with the in situ data. The spectral angle was better than 4.1� for the best cases, in theudspectral range of 450–750 nm. The retrieval of chlorophyll-a (Chl-a) concentration using a popularudsemi-analytical band ratio algorithm for turbid inland waters gave an accuracy of ~16% or 4.4 mg/m3udcompared to retrieval of Chl-a from reflectance measured in situ. Using fixed ATCOR4 processingudparameters for whole images improved Chl-a retrieval results from ~6 mg/m3 difference to referenceudto approximately 2 mg/m3. We conclude that the AisaFENIX sensor, in combination with ATCOR4udin image-driven parametrization, can be successfully used for inland water quality observations.udThis implies that the need for in situ reference measurements is not as strict as has been assumed anduda high degree of automation in processing is possible.