2021ApJ...922...70W

Although microlensing of macrolensed quasars and supernovae provides unique opportunities for several kinds of investigations, it can add unwanted and sometimes substantial noise. While microlensing flux anomalies may be safely ignored for some observations, they severely limit others. "Worst-case" estimates can inform the decision whether or not to undertake an extensive examination of microlensing scenarios. Here, we report "worst-case" microlensing uncertainties for point sources lensed by singular isothermal potentials, parameterized by a convergence equal to the shear and by the stellar fraction. The results can be straightforwardly applied to nonisothermal potentials by utilizing the mass sheet degeneracy. We use microlensing maps to compute the fluctuations in image micromagnifications and estimate the stellar fraction at which the fluctuations are greatest for a given convergence. We find that the worst-case fluctuations happen at the stellar fraction κ_*=1/|µ_macro| . For macrominima, the fluctuations in both magnification and demagnification appear to be bounded (1.5 > Δm > -1.3, where Δm is the magnitude relative to the average macromagnification). Magnifications for macrosaddles are bounded as well (Δm > -1.7). In contrast, demagnifications for macrosaddles appear to have unbounded fluctuations as 1/µ_macro - 0 and κ_* - 0.