Retrieval of color producing agents in Case 2 waters using Landsat 8

Abstract

Abstract New approaches need to be considered to solve the current high demand for color producing agent (CPA) retrievals over inland and coastal waters (Case 2 waters). Traditional retrieval algorithms are known to fail over highly turbid Case 2 waters because they were developed specifically for the open ocean (Case 1 waters). Landsat 8 provides an improved signal-to-noise ratio (SNR) and a new spectral coastal aerosol band in the blue. This additional information provides means to tackle this retrieval endeavor. A look-up-table (LUT) and spectrum-matching methodology was implemented to simultaneously retrieve CPAs, taking advantage of Landsat 8's new features. A LUT of spectral remote-sensing reflectances (Rrs) with different concentrations of CPAs was produced using the in-water radiative transfer model HydroLight. A model-based empirical line method (MoB-ELM) algorithm was developed to atmospherically correct the Landsat 8 imagery and allow direct comparison with the LUT of Rrs. This MoB-ELM atmospheric correction algorithm uses pseudo-invariant features (PIFs) from the image, ground-truth data and the HydroLight model. The retrieval algorithm was applied over three Landsat 8 scenes and shows a normalized root mean squared error (NRMSE) of about 14% for chlorophyll-a, 11% for total suspended solids (TSS), and 7% for colored dissolved organic matter (CDOM) when compared with ground-truth data. The CPA concentration maps exhibit expected trends of low concentrations in clear water and higher concentrations in turbid water. These results demonstrate that the developed algorithm allows the simultaneous mapping of concentration of all CPAs in Case 2 waters and over areas where the traditional algorithms fail or are not available due to spatial resolution. Therefore, this study shows that the Landsat 8 satellite can be utilized over Case 2 waters as long as a careful atmospheric correction is applied.