Abstract
Retrieval of aquatic biogeochemical variables, such as the near-surface concentration of chlorophyll-a (Chla) in inland and coastal waters via remote observations, has long been regarded as a challenging task. This manuscript applies Mixture Density Networks (MDN) that use the visible spectral bands available by the Operational Land Imager (OLI) aboard Landsat-8 to estimate Chla. We utilize a database of co-located in situ radiometric and Chla measurements (N = 4,354), referred to as Type A data, to train and test an MDN model (MDNA). This algorithm’s performance, having been proven for other satellite missions, is further evaluated against other widely used machine learning models (e.g., support vector machines), as well as other domain-specific solutions (OC3), and shown to offer significant advancements in the field. Our performance assessment using a held-out test data set suggests that a 49% (median) accuracy with near-zero bias can be achieved via the MDNA model, offering improvements of 20 to 100% in retrievals with respect to other models. The sensitivity of the MDNA model and benchmarking methods to uncertainties from atmospheric correction (AC) methods, is further quantified through a semi-global matchup dataset (N = 3,337), referred to as Type B data. To tackle the increased uncertainties, alternative MDN models (MDNB) are developed through various features of the Type B data (e.g., Rayleigh-corrected reflectance spectra ρs). Using held-out data, along with spatial and temporal analyses, we demonstrate that these alternative models show promise in enhancing the retrieval accuracy adversely influenced by the AC process. Results lend support for the adoption of MDNB models for regional and potentially global processing of OLI imagery, until a more robust AC method is developed. Index Terms—Chlorophyll-a, coastal water, inland water, Landsat-8, machine learning, ocean color, aquatic remote sensing.
Highlights
Near-surface concentration of chlorophyll-a (Chla), a proxy for phytoplankton biomass, has been observed and quantified in aquatic ecosystems through optical remote sensing for many years (Clarke et al, 1970; Wezernak et al, 1976; Smith and Baker 1982; Gordon et al, 1983; Bukata et al, 1995)
The Mixture Density Networks (MDN) model outperforms other machine learning (ML) models with improvements in error ranging from 30 to 60%, and Multilayer Perceptrons (MLP) ranking as the second-best performer
Given the coarse hyperparameter grid search performed for other ML methods, and the lack of equivalent optimization for MDN parameters, the improvement in performance is even more significant than shown here (Supplementary Appendix C)
Summary
Near-surface concentration of chlorophyll-a (Chla), a proxy for phytoplankton biomass, has been observed and quantified in aquatic ecosystems through optical remote sensing for many years (Clarke et al, 1970; Wezernak et al, 1976; Smith and Baker 1982; Gordon et al, 1983; Bukata et al, 1995). This technique has led to the routine production of Chla distributions for the global oceans for more than two decades. The RE observations, are not available in the suite of measurements made by heritage missions – such as Landsat—which have provided the longest record of Earth observation from space (Goward et al, 2017)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
