Lessons Learned from IPHEx—Challenges to the Representation of Low-Level Aerosol–Cloud–Precipitation Interactions in Models

Differences on the order of 10^oC in surface temperature between adjacent hillslopes tied to mid-day and afternoon low level cloudiness are key to the maintenance of warm-season near-surface microclimates that explain the rich biodiversity of the Southern Appalachian Mountains (SAM), not unlike cloud forests elsewhere from the tropics to the mid-latitudes. In addition to changes in the surface energy budget, persistent low-level cloudiness strongly impacts the diurnal cycle of rainfall through seeder-feeder interactions among layered and stratiform clouds that result in early morning, mid-day and early afternoon rainfall. Using measurements from the Integrated Hydrology and Precipitation Experiment (IPHEx) in 2014, the Advanced Weather Research and Forecasting (WRF) model was used to investigate the Cloud Condensation Nuclei (CCN) sensitivity of simulated summer cloudiness and rainfall by contrasting simulations using "standard" continental (control) and local CCN (IPHEx) spectra. The simulated cloud droplet number concentration using the IPHEx CCN show overall better agreement with airborne observations and better replicate the widespread mid-day low-level cloudiness in the inner Southern Appalachians Mountains(SAM). However, IPHEx simulations exhibit well-defined features of stratiform precipitation with embedded convection and drizzle suppression across the region. The delay of the onset of early afternoon precipitation in IPHEx compared to control simulations is associated with high CCN concentrations that lead to more numerous and smaller cloud droplets. The suppression of warm-rain in the region allows more cloud water to be available that when transported vertically freezes at higher levels releasing latent heat that invigorates convection. In the control simulations, warm rain is produced before precipitating systems with ice content at higher levels develop. Because the summer of 2014 was very dry and atypical with regard to the formation of mid-day low level cloudiness, warm season simulations were also conducted for selected events during 2010 and 2012 using IPHEx and control CCN spectra to examine how regional moisture convergence patterns, topography and aerosol-cloud-precipitation interactions are represented in the model. The results reveal weaknesses in the model simulated vertical structure of clouds and rainfall microphysics in the lower troposphere that result in unrealistic suppression of clouds and consequently rainfall, which in turn strongly impact simulated surface heat fluxes and, therefore, simulated land-atmosphere interactions. A strategy to address these limitations is proposed.