TL;DR
This study leverages seven years of CALIPSO lidar data to provide the most comprehensive, observation-based quantification of trans-Atlantic dust transport to date. By resolving dust in three dimensions, the research offers crucial insights into its seasonal and interannual variability, its relationship with climate indices, and the potential biases in previous satellite-based estimates.
What Problem Does This Paper Address?
The trans-Atlantic transport of mineral dust from North Africa is a critical component of Earth's climate and biogeochemical cycles, impacting human health, ocean fertilization, and atmospheric processes. While satellite observations have long been used to track dust plumes, accurately quantifying the mass flux (the amount of dust transported per unit time) has been challenging. Previous estimates, primarily relying on passive satellite sensors like MODIS, faced limitations due to their inability to fully resolve the vertical distribution of aerosols, especially above clouds, and relied on assumptions about aerosol properties. This study aims to overcome these limitations by using the active sensing capabilities of the CALIPSO lidar to provide a more robust, vertically resolved, and multi-year estimate of dust mass flux.
Methods
The core of this study lies in the sophisticated analysis of CALIPSO's Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) data. The authors employed a 7-year (2007-2013) record of CALIOP's aerosol profiles, focusing on cloud-free (CSA) and above-cloud (ACA) conditions.
Key methodological steps include:
- Dust Identification: Using CALIOP's depolarization ratio measurements, mineral dust was distinguished from other aerosol types (e.g., marine, smoke). This involved defining thresholds for depolarization ratios of dust ($\delta_d$) and non-dust aerosols ($\delta_{nd}$), exploring both "lower-bound" (LDF) and "upper-bound" (UDF) scenarios to capture uncertainty.
- Vertical Profiling: CALIOP's ability to provide vertical profiles of aerosol backscatter and extinction was crucial. This allowed for the calculation of dust extinction profiles.
- Mass Flux Calculation: Dust extinction profiles were converted to dust mass concentration using an assumed dust mass extinction efficiency (MEE). This mass concentration was then combined with zonal wind speed data (from MERRA reanalysis) to calculate dust mass flux rates.
- All-Sky vs. Clear-Sky: The study explicitly addressed the potential bias introduced by using only clear-sky observations. By incorporating both CSA and ACA data, they calculated an "all-sky" dust mass flux and quantified the difference.
- Longitudinal Cross-Sections: Dust transport was analyzed across three key longitudinal cross-sections (15°W, 35°W, and 75°W) to capture dust outflow from North Africa, transport across the Atlantic, and arrival in the Caribbean region.
Datasets
- CALIPSO Lidar Data: The primary dataset used was the CALIPSO version 3, level 2 lidar data, covering the period from December 2006 to November 2013. This included aerosol profiles and cloud layer products.
- MERRA Reanalysis Data: Meteorological data, specifically zonal wind speed, was obtained from the Modern-Era Retrospective Analysis for Research and Applications (MERRA).
- AERONET Ground-Based Data: For validation and comparison, ground-based Aerosol Robotic Network (AERONET) data from sites in Cape Verde and La Parguera were used to compare aerosol optical depth (AOD) and dust optical depth (DOD) trends.
- Climate Indices: Data for the North Atlantic Oscillation (NAO) index and the Sahel Precipitation Index (SPI) were used to investigate interannual variability.
Key Insights and Ideas
- Vertical Resolution Matters: The study highlights how CALIOP's 3D aerosol profiling significantly improves dust transport quantification compared to previous 2D (column-integrated) approaches. This is particularly evident in understanding dust layer height and its seasonal variations.
- Above-Cloud Aerosols (ACA) are Important: The ability to measure aerosols above clouds revealed that ACA can be significant, especially at certain longitudes and seasons, and that neglecting it can lead to biases in flux estimates.
- Dust Separation Uncertainty: The choice of depolarization ratio thresholds for separating dust from non-dust aerosols introduces a notable uncertainty, which increases with distance from the source region.
- Climate Linkages: The strong anti-correlation between annual dust mass flux and the prior-year Sahel Precipitation Index (SPI) suggests SPI could be a useful predictor for dust transport. Similarly, the correlation with the NAO index points to large-scale atmospheric circulation patterns influencing dust transport.
- Seasonal and Meridional Shifts: Dust transport isn't static; it exhibits clear seasonal shifts in its latitudinal distribution, largely driven by the Intertropical Convergence Zone (ITCZ), impacting different regions of the Americas at different times of the year.
Key Results
- Quantified Dust Flux: The 7-year average dust mass flux to 15°W (near North Africa) was estimated at 182 Tg yr⁻¹, decreasing to 132 Tg yr⁻¹ at 35°W and 43 Tg yr⁻¹ at 75°W.
- Seasonal Dominance: Dust transport peaks in JJA (summer) and is lowest in SON (fall).
- Interannual Variability: Significant year-to-year variations in dust flux were observed, with variations up to 86%.
- Clear-Sky Bias: Using only clear-sky observations can overestimate dust mass flux by up to 20% (and even 42% in specific years) at 35°W and 75°W during boreal winter and spring.
- Climate Index Correlations: Annual dust mass flux is highly anti-correlated with prior-year SPI (R² of 0.59-0.86) and positively correlated with DJF NAO index (R of 0.70-0.83).
- Overall Uncertainty: The study estimates the overall known uncertainty in dust mass flux to be ±(45-70)%.
Limitations and Future Work
- Below-Cloud Aerosols: CALIOP cannot detect aerosols below optically thick low-level clouds, meaning the flux estimates do not account for this potentially significant contribution.
- Diurnal Variation: The study primarily uses nighttime CALIOP data for quality reasons. While diurnal variations in dust AOD are found to be minor in transported regions, this assumption could introduce some uncertainty.
- Dust MEE: The assumption of a constant dust mass extinction efficiency (MEE) throughout the transport path might not fully capture real-world changes in particle size distribution.
- Spatial Sampling: CALIOP's discrete sampling might miss some fine-scale variations.
Future work could involve:
- Integrating ground-based lidar data to better characterize below-cloud aerosols.
- More comprehensive analysis of diurnal variations using geostationary satellite data.
- Investigating the impact of varying dust MEE on flux estimates.
- Further exploring the role of meridional transport and its contribution to dust deposition in regions like the Amazon Basin.
Summary updated