Passive microwave radiometers are important remote sensing tools for Earth’s oceans, atmosphere, land and cryosphere. Technologies to improve radiometer spectral resolution over the past decade have been advanced, but spatial resolution has remained stagnant. The models of the Earth system in the meantime are continually evolving to higher spatial resolutions (sub-km scale) with less reliance on parameterizations in favor of physical models surpassing passive MW observational capability. Furthermore, new machine learning based weather models are showing promise by “learning” complex physical processes – breakthroughs may occur when fed with km-resolution microwave observations. Applications include atmospheric 3D thermodynamic processes (e.g. boundary layer temperature and moisture, cloud and precipitation structure) and high-resolution surface imaging (temperature, soil moisture, surface wind vector, ocean salinity, vegetation water content, snow depth). The high spatial resolution will be achieved with large arrays at UHF to Ka-band frequencies and with focal plane arrays at millimeter wave frequencies (>45 GHz). Technology development is focused on thinned synthetic array systems in combination with analog beamforming, and self-calibrating, stable ultra-low noise MMIC RF receiver technology for focal plane arrays.