With hyperspectral imaging, image content can be identified based on fine spectral details related to chemical composition. Immediate applications in smart agriculture and environmental monitoring have the potential for strong societal benefits. However, the technology struggles with the vast amount of data that it produces, in particular when deployed on satellites. The current movement towards increased use of lossy compression is highly risky, because even careful and tedious parameter tuning cannot guarantee that no applications are compromised. We implemented and validated a compression method that simultaneously provides a strong data reduction and preserves analysis results for all possible applications.