Earth Observation through microwave radiometry is particularly useful for various applications, e.g., soil moisture, ocean salinity, or sea ice cover. However, most of the image processing/data analysis techniques aiming to provide automatic measurement from remote sensing data do not rely on any spatial information, similarly to the early years of opti-cal/hyperspectral remote sensing. After more than a decade of research, it has been observed that spatial information can very significantly improve the accuracy of land use/land cover maps. In this context, the goal of this paper is to propose a few insights on how spatial information can benefit to (passive) microwave remote sensing. To do so, we focus here on mathematical morphology and provide some illustrative examples where morphological operators can improve the processing and analysis of microwave radiometric information. Such tools had great influence on multispectral/hyperspectral remote sensing in the past, and are expected to have a similar impact in the microwave field in the future, with the launch of upcoming missions with improved spatial resolution, e.g. SMOS-NEXT.