TL;DR
This study uses 15 years of MODIS satellite data and CAM5 climate model simulations to disentangle the interannual variability and trends of combustion aerosol and dust in major continental outflow regions. It reveals distinct regional trends, with decreasing combustion aerosol over North America and Europe, and increasing trends over South Asia and the Middle East, highlighting the impact of changing emissions.
What Problem Does This Paper Address?
Understanding how aerosols, particularly combustion aerosols (from burning fossil fuels and biomass) and mineral dust, change over time and across continents is crucial for climate modeling, air quality assessment, and understanding their impacts on human health and ecosystems. While satellite data provides broad coverage, distinguishing between different aerosol types and attributing observed trends to specific sources has been a significant challenge. This paper aims to address this by providing a detailed, type-specific analysis of aerosol trends and variability in key outflow regions.
Methods
The study employs a two-pronged approach:
- MODIS Aerosol Retrieval: They utilize the Collection 6 MODIS Dark Target (DT) algorithm's over-ocean aerosol optical depth (AOD) data from both Terra and Aqua satellites. A key methodological step is the reassessment and derivation of characteristic "fine-mode fractions" (FMF) for combustion aerosol, dust, and marine aerosol to accurately separate their contributions to the total AOD.
- CAM5 Climate Model Simulations: They use the Community Atmosphere Model version 5 (CAM5) with source-tagging capabilities. This allows them to track the origin and composition of aerosols, providing a model-based perspective to complement the satellite observations and help interpret the observed trends.
Datasets
- MODIS Collection 6 Aerosol Products: Specifically, the Level 3 daily products (MOD08D3 and MYD08D3) for both Terra and Aqua satellites, covering the period 2003-2017.
- Community Atmosphere Model version 5 (CAM5) simulations: A 37-year simulation (1979-2015) using time-varying emissions and source-tagging capabilities.
- MERRA-2 reanalysis data: Used to nudge the CAM5 model's wind fields.
- CMIP6 emission databases: For monthly anthropogenic and open biomass burning emissions.
Key Insights and Ideas
- Aerosol Type Matters: Simply looking at total AOD can mask important regional differences and trends. Separating combustion aerosol from dust is critical for understanding the drivers of atmospheric change.
- Emissions Drive Trends: The observed trends in combustion aerosol are strongly linked to regional changes in anthropogenic emissions, with decreases in North America and Europe and increases in South Asia and the Middle East.
- Model-Satellite Agreement Varies: While the CAM5 model generally agrees with MODIS on combustion aerosol trends and variability, there are significant discrepancies in dust AOD, suggesting areas for model improvement.
- Terra vs. Aqua Data: The study highlights potential issues with the MODIS Terra sensor's long-term trend, recommending the use of MODIS Aqua data for trend analysis due to its more consistent calibration.
- Episodic Events Challenge Models: Regions dominated by biomass burning smoke or dust show large interannual variability but often lack statistically significant trends, indicating the difficulty in modeling these episodic events.
Key Results
- North America & Europe: Decreasing trends of combustion aerosol (-0.017 to -0.020 per decade) are observed by both MODIS and CAM5, attributed to reduced emissions.
- South Asia & Middle East: Increasing trends of combustion aerosol (+0.017 to +0.036 per decade) are seen, linked to rising anthropogenic emissions.
- Northwestern Pacific: MODIS shows a decreasing trend in combustion aerosol (-0.021 per decade) and dust (-0.012 per decade), which CAM5 does not reproduce.
- Dust Discrepancy: MODIS-derived dust AOD over oceans is significantly higher (about 5 times) than that from CAM5, even after accounting for potential cloud contamination.
- Seasonal Trends: Trends in combustion aerosol show strong seasonal variations, with more pronounced decreases in summer over the North Atlantic and Mediterranean.
Limitations and Future Work
- Dust AOD Discrepancy: The significant difference between MODIS and CAM5 dust AOD remains a challenge, with potential contributions from MODIS cloud contamination and model underestimation. Further investigation is needed.
- Data Record Length: While 15 years of data are deemed adequate for regions with consistent emission trends, longer records may be necessary for regions with non-monotonic emission changes or where trends are entangled with meteorological variability.
- Aerosol Categorization: The broad categorization of "combustion aerosol" does not distinguish between fossil fuel and biomass burning sources, limiting a more detailed understanding. Future work could benefit from satellite retrievals of aerosol chemical composition.
- Model Sensitivity Experiments: The authors suggest that well-designed model sensitivity experiments are needed to disentangle the roles of emissions and meteorology in aerosol variability and trends.
- VIIRS Data: The study recommends that similar assessments of characteristic FMF be conducted for VIIRS data to ensure continuity of aerosol records.
Summary updated