The influence of simulated surface dust lofting and atmospheric loading on radiative forcing

Abstract

Abstract. This high-resolution numerical modeling study investigates the\npotential range of impact of surface-lofted dust aerosols on the mean\nradiative fluxes and temperature changes associated with a dust-lofting\nepisode over the Arabian Peninsula (2–5\xa0August\xa02016). Assessing the\npotential for lofted dust to impact the radiation budget and temperature\nresponse in regions of the world that are prone to intense dust storms is\nimportant due to the impact of such temperature perturbations on thermally\ndriven mesoscale circulations such as sea breezes and convective outflows.\nAs such, sensitivity simulations using various specifications of the dust-erodible fraction were performed using two high-resolution mesoscale models\nthat use similar dust-lofting physics based on threshold friction wind\nvelocity and soil characteristics. The dust-erodible fraction, which\nrepresents the fraction (0.0 to 1.0) of surface soil that could be\nmechanically lifted by the wind and controls the location and magnitude of\nsurface dust flux, was varied for three experiments with each model. The\n“Idealized” experiments, which used an erodible fraction of 1.0 over all\nland grid cells, represent the upper limit on dust lofting within each\nmodeling framework, the “Ginoux” experiments used a 1 ∘ resolution,\nspatially varying erodible fraction dataset based on topographic\ndepressions, and the “Walker” experiments used satellite-identified, 1\u2009km\nresolution data with known lofting locations given an erodible fraction of\n1.0. These simulations were compared with a “No-Dust” experiment in which no\ndust aerosols were permitted. The use of erodible fraction databases in the\nGinoux and Walker simulations produced similar dust loading which was more\nrealistic than that produced in the Idealized lofting simulations. Idealized\nlofting in this case study generated unrealistically large amounts of dust\ncompared with observations of aerosol optical depth (AOD) due to the lack of\nlocational constraints. Generally, the simulations with enhanced dust mass\nvia surface lofting experienced reductions in daytime insolation due to\naerosol scattering effects as well as reductions in nighttime radiative\ncooling due to aerosol absorption effects. These radiative responses were\nmagnified with increasing amounts of dust loading. In the Idealized\nsimulation with extreme (AOD >\u20095) dust amounts, these\nradiative responses suppressed the diurnal temperature range. In the Ginoux\nand Walker simulations with moderate (AOD ∼1 –3) amounts\nof lofted dust, the presence of dust still strongly impacted the radiative\nfluxes but only marginally modified the low-level temperature. The\ndust-induced near-surface temperature change was limited due to competing\nthermal responses to changes in the net radiative fluxes and the dust-layer\nradiative heating rates. Compared to the Ginoux simulation, the use of\nincreased resolution in dust-erodible fraction inventories in the Walker\nsimulations led to enhanced fine-scale horizontal variability in lofted dust\nand a modest increase in the mean dust concentration profile and\nradiative or thermal responses. This study discusses the utility of using\nhigh-resolution dust source databases for simulating lofted dust, the need\nfor greater spatial coverage of in situ aerosol observations in dust-prone\nregions, the impacts of dust on the local radiation budget and surface\nthermal conditions, and the potential dust radiative impacts on thermally\ndriven mesoscale features.