Source: Journal of Advances in Modeling Earth Systems
In nature, the distinction between rain and cloud droplets is just that rain droplets are large enough that their terminal velocity exceeds the updraft velocity, so they fall to the ground, whereas cloud droplets are so small that they remain suspended. Essentially all climate and weather forecasting models introduce artificial gaps into the continuous spectrum of droplets, calling the larger rain, the smaller cloud droplets, and then model the conversion from one to the other. Modeling this conversion process is fraught and has given rise to large errors in simulating rain.
Igel et al.  show that representing the droplet size spectrum as the continuum it is in nature leads to substantial improvements in simulations of rain in shallow clouds, even when the continuum is represented by only a few statistical moments so that the computational expense of their new method does not exceed that of traditional methods with separate rain and cloud categories. This is another example of how being more faithful to physical processes and avoiding artificial discontinuities in models can improve critical parameterizations of small-scale processes—and especially microphysical processes that will remain out of reach for direct simulation for the foreseeable future.
Citation: Igel, A. L., Morrison, H., Santos, S. P., & van Lier-Walqui, M. (2022). Limitations of separate cloud and rain categories in parameterizing collision-coalescence for bulk microphysics schemes. Journal of Advances in Modeling Earth Systems, 14, e2022MS003039. https://doi.org/10.1029/2022MS003039
—Tapio Schneider, Editor, Journal of Advances in Modeling Earth Systems