Lidar as an Evaluation Tool for a Newly Developed Aerosol Satellite Radiance Assimilation System

Operational numerical weather prediction systems are critically reliant on hyperspectral-IR data assimilation, but they remain susceptible to significant biases in the absent of atmospheric correction, particularly aerosols and clouds. For the most part, aerosol “contaminated” fields are rejected before getting assimilated, significantly reducing the amount of satellite data that is ingested in critical weather areas. With this in mind, the U.S Naval Research Lab (NRL) has developed the capability of assimilating aerosol-laden satellite radiances into operational DA. The newly introduced hyperspectral radiance signatures from the four major hyperspectral platforms used for Numerical Weather Prediction (IASI, AIRS, CrIS and GeoCSR), utilize vertical distributions from the NAVY Aerosol Analysis Prediction System (NAAPS) and optical properties from the optical properties of aerosols and clouds (OPAC). However, the accuracy of brightness temperature representation in radiative transfer models (and therefore, data assimilations and forecast outputs), is directly correlated to aerosol loading and with correctly prescribing aerosols within the atmospheric column. Validation is necessary and lidar datasets are ideal to provide corroboration for the newly developed system.

Here, we evaluate the performance and robustness of the new NRL’s “ AERO-DA” system against global lidar observations iterating two case studies over 3-6 month periods (hurricane seasons 2016-2017). Inter-comparisons with aerosol, temperature, and relative humidity observations are done by collocating CALIPSO and MPL profiles with those parameters used as input (and output) in AERO-DA. Our focus is over the primary global oceanic regions (tropical Atlantic and Caribbean, northern Indian Ocean and western Pacific Ocean). We also demonstrate how lidar can potentially improve the screening of observations impacted by aerosols and clouds in our DA system. Finally, we show how the both the modeling and the observation communities can synergistically exploit the full value of incoming satellite data, and prepare for the next-generation hyperspectral sounders using current lidar platforms.