Quantifying aerosol indirect effects on ice and mixed-phase clouds remains a challenging research topic. This is partly due to the large spatial heterogeneities of these clouds, as well as various key factors that can affect cloud microphysical properties simultaneously, such as thermodynamic effect, dynamic effect, and aerosol indirect effect. Previous studies often focused on one factor at a time, yet a comprehensive analysis of these key factors is needed. This work developed a synthesized in-situ observation database of ice and mixed-phase clouds using eight NSF flight campaigns from 87°N to 75°S. Statistical distributions of cloud microphysical properties are quantified for the cirrus cloud regime (≤ -40°C) and the mixed-phase cloud regime (-40°C to 0°C). A method is developed to isolate effects of thermodynamic (i.e., temperature and relative humidity) and dynamic (vertical velocity) conditions before quantifying aerosol indirect effects on cirrus clouds (Patnaude and Diao, 2020, GRL). Observations show positive correlations of ice/liquid number concentrations (Nice and Nliq) and ice/liquid water content (IWC and LWC) with respect to aerosol number concentrations. Significant increases of IWC and Nice are seen when aerosol concentrations are higher than their average conditions by a factor of 3 to 10, highlighting a non-monotonic and non-linear response of cirrus cloud microphysical properties to aerosol concentrations. When comparing with climate model simulations by the NCAR Community Atmosphere Model (CAM) and DOE Energy Exascale Earth System Model (E3SM), the simulations show weaker correlations for Nice and Nliq and no obvious correlations for IWC and LWC with respect to aerosol concentrations (Patnaude et al., 2020; Yang et al., 2020). These results suggest that aerosol indirect effects on ice and mixed-phase clouds are underestimated in these climate model simulations.