maria: A novel simulator for forecasting (sub-)mm observations

Millimeter-wave single-dish telescopes offer two key advantages compared to interferometers: they can efficiently map larger portions of the sky, and they can recover larger spatial scales. Nonetheless, fluctuations in the atmosphere limit the accurate retrieval of signals from astronomical sources. To efficiently reduce atmospheric noise and filtering effects in current and future facilities, we introduce {\tt maria}, a versatile and user-friendly multi-purpose telescope simulator that optimizes scanning strategies and instrument designs, produces synthetic time-ordered data, time streams, and maps from hydrodynamical simulations, thereby enabling a fair comparison between theory and observations. Each mock observatory scans through the atmosphere in a configurable pattern over the celestial object. We generate evolving and location-and-time-specific weather for each of the fiducial sites using a combination of satellite and ground-based measurements. While {\tt maria} is a generic virtual telescope, this study specifically focuses on mimicking broadband bolometers observing at 100 GHz. We compare the mock time streams with real MUSTANG-2 observations and find that they are quantitatively similar by conducting a k-sample Anderson-Darling test resulting in a p-value of p<0.001. Subsequently, we image the TODs to create noise maps and realistic mock observations of clusters of galaxies for both MUSTANG-2 and an instrument concept for the 50m Atacama Large Aperture Submillimeter Telescope (AtLAST). Furthermore, using {\tt maria}, we find that a 50m dish provides the highest levels of correlation of atmospheric signals across adjacent detectors compared to smaller apertures (e.g., 42-cm and 6-m), facilitating removal of atmospheric signal on large scales.