A neural network estimation of phytoplankton, aerosol and cloud properties using CALIPSO measurements

The Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) onboard the CALIPSO satellite has been operating since June 2006 and is expected to continue operating beyond 2018. Over the life of the CALIPSO mission, the stability of CALIOP's calibrated atmospheric backscatter measurements has remained within 1%. CALIOP's highly stable dual polarization measurements can be used for studying changes in aerosols, cloud microphysical properties, phytoplankton backscatter and carbon biomass (Behrenfeld et al., 2013; Lu et al., 2014; Hu et al., 2015), and their interactions.

While studying inter-annual variations in CALIOP's 532-nm cross-polarization backscatter measurements, we found a clear temporal correlation between water clouds and the ocean subsurface in middle and high latitude regions (e.g., in the Southern Oceans and the storm track polar regions in the northern hemisphere). This finding leads us to pose the following question: does this correlation imply a potential connection between ocean biogeochemistry and cloud microphysical properties at high latitudes? To provide an answer, we started developing innovative retrieval algorithms for a comprehensive analysis of the full 12 years of CALIOP observations of marine aerosols, phytoplankton, and clouds. The new lidar retrieval method and data analysis techniques include,

- Phytoplankton particulate backscatter coefficient that does not rely on collocated MODIS observation;

- Developing and improving innovative retrieval techniques for water cloud droplet size and droplet number concentration and phytoplankton particulate backscatter coefficients;

- Improving an innovative retrieval technique for marine aerosol lidar ratio.

We will introduce the retrieval concepts, retrieved physic properties and their potential applications in studying air-sea interactions.