A Scheme of Atmospheric Correction with A Revised POLYMER model Using Same-day Observations

Abstract:

Based on POLYMER (POLYnomial based approach applied to MERIS data) atmospheric correction model, we present a novel retrieval scheme (TAACA) that determines oceanic and atmospheric optical properties simultaneously based on two same-day satellite images with different sun-sensor viewing geometries. First, the performance of TAACA is validated with a synthetic dataset, and it is found that the retrieved remote-sensing reflectance (R_rs) matched very well (R²≥0.98) with the simulated R_rs, where the unbiased root mean square error (uRMSE) for both non-absorbing and strongly absorbing aerosol cases is ~12.2%. When the synthetic data is processed by POLYMER, the R² and uRMSE values are ~0.99 and ~7.5%, respectively, for non-absorbing aerosols, but they are ~0.92 and ~39.5% for strongly absorbing aerosols. TAACA is further evaluated using co-located VIIRS measurements over the Massachusetts Bay, and found that the retrieved R_rs from VIIRS agrees with in-situ measurements within ~20.4% at the visible wavelengths. In comparison, NOAA’s operational atmospheric correction algorithm (MSL12 – v1.2) resulted in a difference of 164.9% and 33.1% at 410 nm and 443 nm, respectively, for these measurements. The R_rs products derived from POLYMER also show large deviations from in-situ measurements. It is envisioned that this new scheme provides a promising option of atmospheric correction in coastal waters when there is the presence of strongly absorbing aerosols.