研究目的
This study focuses on the comparison between the standard and regional bio-optical algorithms with in situ TChla, (referred to as TChlaREF). The bio-optical regional products are computed with a multilayer perceptron (MLP) neural network.
研究成果
The study demonstrates the importance of regression schemes that can ef?ciently use Rrs values at individual wavelengths, as in the regional MLP algorithms adopted here. It also shows the feasibility and relevance of using regional algorithms for timely evaluation of operational space mission deliverables as a cost-effective addition to the direct matchup analyses of primary radiometric data and derived bio-geophysical quantities.
研究不足
The study reveals that differences between standard and regional API1 estimates vary among the sampling stations, with station A more affected by adjacency due to its proximity to the coast. The need to improve the atmospheric correction and reduce the larger Rrs uncertainties in the blue and red spectral regions is also noted.
