Importance of high-latitude dust in the Arctic
Submitter:
Liu, Xiaohong — Texas A&M University
Area of research:
Cloud-Aerosol-Precipitation Interactions
Journal Reference:
Science
Dust particles, serving as ice-nucleating particles (INPs), may impact the arctic surface energy budget and regional climate by modulating arctic mixed-phase cloud properties. Although most of the dust aerosols were thought to originate from arid regions in the low latitudes (e.g., North Africa and Asia), it is increasingly recognized that a large amount of dust is also produced by local arctic sources (i.e., high-latitude dust, HLD). HLD was found to nucleate ice as efficiently as, or even more efficiently than, low-latitude dust (LLD). However, HLD sources are largely underrepresented or even omitted in most global models and their impacts on arctic clouds are overlooked. In this work, we perform a modeling study to quantify the importance of HLD in the Arctic.
Impact
Using a global climate model, this study presents clear evidence that HLD plays a significant role in arctic clouds. We found that HLD has a large contribution to the dust loading in the arctic lower troposphere, while the LLD dominates the upper troposphere. Also, the modeled INP concentrations show better agreement with INP measurements in the Arctic when including HLD INPs, which indicates the contribution of HLD to the arctic INP population. We find that the HLD INPs result in a net cooling effect on the arctic surface by glaciating mixed-phase clouds, the magnitude of which is larger than the radiative effect induced by LLD INPs from North African and East Asian sources. Considering the climate impacts of HLD in the Arctic will be more important given a warming climate in the future, wherein reduction in snow coverage and more exposure of dry land in the Arctic may lead to increased HLD emissions.
Summary
In this study, we evaluate the contribution to arctic dust loading and INP population from HLD and six LLD source regions by implementing a source-tagging technique for dust aerosols in version 1 of the U.S. Department of Energy’s Energy Exascale Earth System Model (E3SMv1). Our results show that HLD is responsible for 30.7% of the total dust burden in the Arctic on the annual mean, whereas LLD from Asia and North Africa contributes 44.2% and 24.2 %, respectively. Due to its limited vertical transport as a result of stable boundary layers, HLD contributes more in the lower troposphere, especially in boreal summer and autumn when the HLD emissions are stronger. LLD from North Africa and East Asia dominates the dust loading in the upper troposphere with peak contributions in boreal spring and winter. The modeled INP concentrations show better agreement with both ground and aircraft INP measurements in the Arctic when including HLD INPs. The HLD INPs are found to induce a net cooling effect (-0.24 W m-2 above 60oN) on the arctic surface downwelling radiative flux by changing the cloud phase of arctic mixed-phase clouds. The magnitude of this cooling is larger than that induced by North African and East Asian dust (0.08 and -0.06 W m-2, respectively), mainly due to different seasonalities of HLD and LLD.
Figure 1. Comparison of predicted versus observed INP concentrations in the Arctic. The predicted INP concentrations are derived from (a) LLD using classical nucleation theory (CNT), (b) LLD and HLD both using CNT, (c) LLD using CNT and HLD using Sanchez-Marroquin et al. (2020; SM20), (d) LLD using DeMott et al. (2015; D15), (e) LLD and HLD both using D15, (f) LLD using D15 and HLD using SM20, (g) black carbon (BC) using Schill et al. (2020; Sc20), and (h) sea spray aerosol (SSA) using McCluskey et al. (2018; M18). SSA includes both marine organic aerosol and sea salt. Nine INP data sets are classified by symbols A to I, the color of which represents the temperature reported in the observations. The modeled INP concentrations are diagnosed using the observed temperatures and monthly averaged aerosol properties of the corresponding month from the year 2007 to 2011. The INP concentrations for CNT are defined as the CNT immersion freezing rate integrated by 10 s, following Hoose et al. (2010) and Wang et al. (2014). The solid line in each panel represents 1:1 comparison, while dashed lines outline 1 order of magnitude differences. From journal.