Air pollution is responsible for ~7 million premature deaths every year. Past and current satellite missions in low Earth orbit have characterized air pollution distributions and established trends at fixed local solar times but in most cases cannot resolve individual emission sources without statistical post-analysis. Current and planned geostationary satellites will add diurnal information but lack global coverage. The Decadal Survey calls for a robust, comprehensive observing strategy for the spatial distribution of air pollution at high spatial, high temporal resolution. This will not be possible in a sustainable way without technological advancements. The Compact Hyperspectral Air Pollution Sensor (CHAPS) is a hyperspectral imager (HSI) using free-form optics in a form factor suitable for accommodation on a small satellite or hosted payload. CHAPS will make measurements of air pollution at unprecedented spatial resolution from low Earth orbit (1 x 1 km2) and will characterize, quantify, and monitor emissions from urban areas, power plants, and other anthropogenic activities. The compact size and relatively lower cost of CHAPS makes a constellation feasible for the first time, with unprecedented spatiotemporal sampling of global point pollution sources. A CHAPS constellation represents a new observing system making science-quality measurements of air pollution, meeting new Decadal Survey requirements. The objective of the CHAPS–Demonstrator (CHAPS-D) IIP project is the airborne demonstration of a CHAPS prototype instrument. CHAPS derives heritage from the TROPOspheric Monitoring Instrument (TROPOMI) on the Sentinel-5 Precursor, which uses a free-form mirror telescope. Free-form optics is an emerging technology with potentially huge advantages over traditional optical designs, including fewer optical surfaces, less mass and volume, and improved image quality. The free-form optics design demonstrated by the CHAPS D IIP will be generalizable to other wavelengths between 270 and 2400 nm, making it applicable to a wide variety of Earth science problems, including public health, atmospheric composition, surface biology and geology, land use/agriculture, marine and terrestrial ecosystems, the cryosphere, volcanic eruptions, and natural disaster response. As a case study, we will focus this IIP project on the measurement of nitrogen dioxide (NO2). NO2 is a primary ingredient of air pollution, as it is a toxic gas at high concentrations, a marker for combustion-related pollutants and co-emitted air toxins, and the main precursor of tropospheric ozone and nitrate aerosols. CHAPS-D will combine a radiometrically calibrated HSI (300–500 nm @ 0.5-nm resolution) with associated detector and payload electronics. It will be as close to the ultimate space design as feasible within the scope of the IIP. For example, we will impose the design constraints for the payload of a 6U CubeSat. With these constraints in mind, we will introduce new technologies for on-board instrument calibration. With new, innovative passive metrology, we will be able to constantly monitor the instrument for continuous on-board correction of the measurement data. The project total period of performance is from January 2020 through December 2022. The first project year focuses on the instrument design and the fabrication of the free-form optics, detector, and electronics. The second year will see the integration of the instrument, laboratory calibration, and ground-based (zenith-sky) measurements. The third year features a series of aircraft flights, where we will validate the performance of CHAPS-D, retrieve NO2 vertical column densities from measured radiance spectra, and compare these retrievals with extant ground-based, airborne, and operational space-based NO2 products. This will raise the CHAPS-D TRL from 2 to 5, preparing the compact hyperspectral imaging technology to tackle numerous Earth science objectives.