Aerosol-Weather-Climate Group

Physical Drivers of Sand/Dust Storms

Dust event location, timing, duration, and intensity are collectively determined by three broad drivers, each represented by a number of processes operating across different spatial and temporal scales:

Sediment Supply: the abundance of fine soil material, generated via weathering, fluvial, and/or aeolian processes on medium- to long-term timescales (e.g., decades to millennia).
Sediment Erodibility: the minimum wind speed required to initiate dust mobilization, primarily modulated by inter-annual to decadal changes in hydroclimate conditions and land use practices.
Wind Erosivity: dominated by rare and often intermittent, high wind events, which may last from minutes to days, generated by various meteorological processes.

Following this analytic paradigm, we diagnosed the two extreme dust events in Central Asia, including an intense salt dust storm from the dried seabed of the former Aral Sea, and the worst dust storm ever recorded in Uzbekistan's history.

Conceptual Framework of Dust Emission Drivers

Role of Atmospheric Blocking in Compound Cold Air and Dust Outbreaks

Atmospheirc blocks are long-lasting, quasi-stationary tropospheric flow patterns associated with an interruption or deceleration of the zonal westerly flow, resulting in a range of surface weather extremes in local and adjacent regions. Blocking systems in the Atlantic-Europe sector play an important role in producing compound cold air and dust outbreaks over Asia. Particularly, blocking in Northern Europe or West Siberia fosters the deep equatorward penetration of polar airmasses, rapid anticyclogenesis, and intense northerly flows capable of dust uplifting from dry, barren surfaces. These dust events are often accompanied by compound weather hazards, such as damaging winds and cold extreme.

In a paper published in Environmental Research Letters, we uncovered the connection between Uzbekistan's historic dust storm in 2021 and upper-tropospheric flow anomalies in upstream areas, including Greenland Blocking, recurrent transient Rossby wave packets, Ural Blocking, and the equatorward excursion of a tropopause polar vortex.

In a forthcoming paper, we will examine the role of Scandinavian Blocking in the spatially compounding European heatwave and Asian cold air and dust outbreaks in summer 2018.

Aerosol Remote Sensing Over Dryland Regions

Satellite remote sensing is the most practical means for monitoring large-scale aerosol events on a daily basis. Drylands (including deserts) present a very challenging environment for spaceborne aerosol observations, due to difficulty in isolating surface contributions from aerosol signals in satellite radiance measurements. This limitation affects passive techniques based on either scattered radiation in the UV and visible-NIR or emitted radiation in the thermal IR. We apply a synergistic, multi-sensor approach and exploit both the heritage and emerging satellite aerosol records to detect the abundance, vertical height, and particle properties of airborne dust, and to infer the locations and dynamics of dust sources.

We conducted a detailed case analysis of multisensor ultraviolet aerosol index (UVAI), mid-visible and infrared aerosol optical depth (AOD), and aerosol layer height (ALH) products, revealing the diversity, inconsistencies, and biases of satellite aerosol algorithms in characterizing salt-rich dust. This paper is now in press.

We have generated global, daily dust AOD records from MODIS for the past 20 years. Ongoing effort is to apply the long-term records for understanding dust source dynamics and drivers.

Development of a Global Unified Dust-Climate Data Record

Currently, dust models rely on measurements far away from the source area for model tuning and evaluation. For dust sources located in the deep interior of continents, aerosol monitoring sites (e.g. AERONET) are scare, while satellite aerosol retrievals are either too sparse or uncertain. This has greatly limited our understanding of dust emission dynamics and quantification of dust model uncertainties and errors.

This work aims to develop a unified Dust-Climate Data Record (DCDR) by merging quality-controlled dust event reports within or near dust source regions, with colocated data on the meteorological, hydroclimate, and geomorphological drivers. DCDR covers the major dust sources around the world, and opens up new opportunities to understand dust emission dynamics, evaluate dust model performance, cross-verify satellite and ground-based dust records, and assess the predictability of dust events.

An initial version of DCDR was used in a global analysis of dust trends, published in the Journal of Geophysical Research Atmospheres. This version can be downloaded from Zenodo.

DCDR is being applied to characterize dust emission dynamics and model errors in predicting dust events. New DCDR versions will be released to the public.

Improving Dust Emission Model Representation and Bias Characterization

Dust emission is arguably the most uncertain component of the dust life cycle, largely due to incomplete or inaccurate physical understanding and model representations. We developed a regional coupled dust model system by incorporating three dust emission schemes into a common host model (WRF) to facilitate the uncertainty analysis. The choice of dust scheme and soil size distribution data contributes to great uncertainties in the interannual variability and anthropogenic fraction of dust emissions.

Our current research includes:

The relative importance of physical drivers in modulating the dust variability in global models.
Application of AI/ML techniques to characterize model errors in predicting dust emission events.

Health Effects of Particulate Pollution

PM2.5 is a leading cause of global disease burden and premature death, especially in low- and middle-income countries. Recently, we performed an assessment of the short-term exposure risk of dust and fine aerosol PM2.5 and the associated premature mortality burden in Central Asia. The daily exposure to unhealthy PM2.5 levels varies from 124 (Kazakhstan) to 251 (Tajikistan) days in 2022. Almaty faced the worst exposure risk to both unhealthy and hazardous PM2.5 levels. An estimated 5074 (95% CI: 3428-6728) premature deaths were attributable to short-term PM2.5 exposure annually across Central Asia.

This study has been published at Science of The Total Environment.

Page updated

Google Sites

Report abuse