WATERHYPERNET: a prototype network of automated in situ measurements of hyperspectral water reflectance for satellite validation and water quality monitoring

Abstract

This paper describes a prototype networkof automated in situ measurements of hyperspectralwater reflectance suitablefor satellite validation and water quality monitoring. Radiometric validationof satellite-derived water reflectance is essentialto ensure that only reliable data, e.g., for estimating water quality parameterssuch as chlorophyll a concentration, reach end-users. Analysis of thedifferences between satellite and in situ water reflectancemeasurements, particularly unmasked outliers, can provide recommendations onwhere satellite data processing algorithms need to be improved. In a massivelymulti-mission context, including Newspace constellations, hyperspectralmissions and missions with broad spectral bands not designed for “water colour”, the advantage of hyperspectralover multispectral in situ measurements is clear. Two hyperspectral measurementsystems, PANTHYR (based on the mature TRIOS/RAMSES radiometer) and HYPSTAR® (a newly designedradiometer), have been integrated here in theWATERHYPERNET network withSI-traceable calibration and characterisation. The systems have common dataacquisition protocol, data processing and quality control. The choice ofvalidation site and viewing geometry and installation considerations aredescribed in detail. Three demonstration cases are described: 1. PANTHYR datafrom two sites are used to validate Sentinel-2/MSI (A&B); 2. HYPSTAR® data at six sitesare used to validate Sentinel-3/OLCI (A&B); 3. PANTHYR and HYPSTAR® data in Belgian NorthSea waters are used to monitor phytoplankton parameters, including Phaeocystis globosa, over two 5 monthperiods. Conclusion are drawn regarding the quality of Sentinel-2/MSI andSentinel-3/OLCI data, including indications where improvements could be made.For example, a positive bias (mean difference) is found for ACOLITE_DSFprocessing of Sentinel-2 in clear waters (Acqua Alta) and clues are provided onhow to improve this processing. The utility of these in situ measurements, even without accompanyinghyperspectral satellite data, is demonstrated for phytoplankton monitoring. Thefuture evolution of the WATERHYPERNET network is outlined, includinggeographical expansion, improvements to hardware reliability and to themeasurement method (including uncertainty estimation) and plans for dailydistribution of near real-time data.