Carbon dioxide (CO2) emissions from strong point sources account for a significant proportion of the global greenhouse gas emissions, and their associated uncertainties in bottom-up estimates remain substantial. Imaging spectrometers provide a capability to monitor large point source CO2 emissions and help reduce the uncertainties. In this study, we assess the capability of an airborne monitoring system with temporally sparse observations to constrain annual emissions at both facility and regional scales. We use observations of power plant emissions from 2022 and 2023 and compare the derived emission rates at facility scale to in-stack emission observations across the United States. We show that CO2 concentration enhancements retrieved using a log-normal matched filter are suitable for CO2 quantification, achieving low bias and uncertainty in estimated emission rates. We find that annual emissions at the regional scale can be effectively constrained by offsetting errors identified at the facility scale, with a 30% uncertainty.