Susceptibility of aerosol optical thickness retrievals to thin cirrus contamination during the BASE‐ASIA campaign

[1] We used a combination of ground measurements (Aerosol Robotic Network, AERONET; Micro-Pulse Lidar Network, MPLNET) and satellite data (Moderate Resolution Imaging Spectroradiometer, MODIS; Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation, CALIPSO) to examine the susceptibility of ground and satellite aerosol retrievals to thin cirrus contamination at Phimai, Thailand (102.56°E, 15.18°N, also known as Pimai), during the Biomass-burning Aerosols in South East-Asia: Smoke Impact Assessment (BASE-ASIA) campaign (February–May 2006). Using the strengths of spaceborne or ground lidars to detect cirrus clouds, we conducted statistical analyses for four different scenarios: MPLNET versus AERONET, MPLNET versus MODIS, CALIPSO versus AERONET, and CALIPSO versus MODIS. Cirrus identifications from MPLNET or CALIPSO were paired up with concurrent aerosol optical thickness (AOT) measurements from AERONET or MODIS. Results from the BASE-ASIA campaign suggest that current operational AERONET and MODIS AOT products are influenced by thin cirrus contamination featuring strong seasonality. Concurrent AERONET and MPLNET observations indicate that additional thin cirrus screening changes AOT monthly means by 5%, with 20% of the AERONET aerosol data at Phimai being cirrus contaminated in boreal spring. From noncirrus cases to cirrus-contaminated cases, AERONET AOT increases along with larger particle sizes. We further evaluated the performance of eight MODIS-derived cirrus screening parameters for their effectiveness on thin cirrus screening: apparent reflectance at 1.38 μm (R1.38), cirrus reflectance at 0.66 μm (CR0.66), CR0.66 cirrus flag, reflectance ratio between 1.38 μm and 0.66 μm (RR1.38/0.66), reflectance ratio between 1.38 μm and 1.24 μm (RR1.38/1.24), brightness temperature difference between 8.6 μm and 11 μm (BTD8.6–11), brightness temperature difference between 11 μm and 12 μm (BTD11–12), and cloud phase infrared approach. Correlation analysis with the MPLNET cirrus flag indicates that RR1.38/0.66 is slightly preferable for high thin cirrus screening for the AERONET AOT measurements. The quantitative findings from this study suggest particular caution and careful evaluation of thin cirrus contamination in the satellite and ground AOT measurements before they are used for aerosol-related climatic forcing studies.