Retrieving MODIS AOD over turbid coastal waters: improvements in spatial resolution and masks for land surfaces and clouds

Abstract:

High spatial resolution Aerosol Optical Depth (AOD) retrievals over coastal waters are useful for investigating both health and visual air quality over coastal region, where more than 60% of population live. However, AOD retrievals over turbid coastal waters are unavailable in the operational Moderate Resolution Imaging Spectroradiometer (MODIS) AOD products. Retrieving AOD over turbid coastal water surface is a challenge for both ocean color and atmospheric aerosol remote sensing communities due to three principle reasons: (a) The high water-leaving radiance caused by sediments in the water could be mis-interpreted as elevated aerosol loadings if ocean surfaces were assumed to be dark; (b) the widely-used cloud mask algorithm is based on clouds higher spatial variability than aerosols and inhomogeneous coastal water turbidity can lead to be misclassified as clouds; (c) land or land-contaminated pixels maybe recognized as water pixels over coastal region, which will lead to abnormal high AOD retrievals, as the high land surface reflectance violates the dark-surface assumption for water surface. In this presentation, we will present new improvements of the past work in which the MODIS 2.1 µm TOA reflectance was used to retrieve AOD over turbid coastal waters. At the 2.1 mm band, water-leaving radiances is negligible regardless of surface water turbidity. New improvements include retrieval of AOD at fine spatial resolution of 1 km and enhanced masks for clouds and land surfaces. We show that using TOA reflectance at 2.1 µm in the cloud mask’s smoothness test successfully avoid misclassifying clear-sky pixels over turbid coastal waters as clouds. Additionally, the 2.1 µm smooth test facilitates detecting land or land-contaminated pixels around coastline as land surface reflectance is much larger than water. Finally, the 1-km coastal water AOD retrievals are well validated with measurements from Maritime Aerosol Network (MAN).