Millimeter-wave telescopes are limited by correlated noise in the celestial signals they can observe at all but the smallest angular scales. Accurate science forecasting for a given instrument depends on our understanding of precisely how well we can recover these scales, which in turn depends on both the properties of the instrument and the mapping algorithm. Accurately simulating correlated noise as seen by a receiver is then a crucial step that informs both the design of the instrument and the design of the mapmaker.
I present maria, a user-friendly Python library for the soup-to-nuts simulation of astronomical observations with realistic correlated atmospheric and instrument noise. I discuss the current use cases of maria, which include observations of the cosmic microwave background, Sunyaev–Zeldovich galaxy clusters, and dynamic solar physics. I propose the future of development, which includes supporting observations with interferometric and heterodyne receivers.
