An assessment of Landsat-8 atmospheric correction schemes and remote sensing reflectance products in coral reefs and coastal turbid waters

The Operational Land Imager (OLI) onboard Landsat-8 satellite can provide remote sensing reflectance (R-rs) of aquatic environments with high spatial resolution (30 m), allowing for benthic habitat mapping and monitoring of bathymetry and water column optical properties. To facilitate these applications, accurate sensor-derived R-rs is required. In this study, we assess atmospheric correction schemes, including NASA's NIR-SWIR approach, Acolite's NIR and SWIR approaches and the cloud-shadow approach. We provide the first comprehensive evaluation for Landsat-8 R-rs retrievals in optically shallow coral reefs, along with an investigation of Landsat-8 R-rs products in a temperate turbid embayment. The obtained Landsat-8 R-rs data products are evaluated with concurrent in situ hyperspectral R-rs measurements. Our analyses show that the NASA and the cloud-shadow approaches generated reliable R-rs products across shallow coral reefs and optically deep waters. This evaluation suggests that high quality R-rs products are achievable from the Landsat-8 satellite in optically shallow environments, which supports further application of Landsat-8 type measurements for coral reef studies.