Mapping depth-integrated microphytobenthic biomass on an estuarine tidal flat using Sentinel satellite data

Abstract

Microphytobenthos (MPB) are central to benthic tidal flat ecological networks. Large-scale information on total MPB biomass is difficult to obtain from traditional in-situ measurements. Here, we assessed the effectiveness of using surface sediment properties as proxies for predicting the total (depth-integrated) MPB biomass with Sentinel satellite data. First, the best subset regression was applied to the in-situ data to determine the properties that best predict the biomass decay rate with depth (sampling every cm up to 10 cm depth). Then, data from a controlled laboratory experiment were used to analyse the spectral response to different sediment properties (i.e., variations in sediment grain size, organic matter and water content). Subsequently, an algorithm was developed to obtain the spatial distribution of the sediment properties and the depth-integrated total MPB biomass from remote sensing data. Finally, we presented a case study in which the seasonal dynamics of total (depth-integrated) MPB biomass were obtained from Sentinel-2 Multispectral Instrument (MSI) satellite data using the Google Earth Engine (GEE) platform. The results showed that (1) the vertical distribution of MPB biomass on a tidal flat could be predicted by the surface MPB biomass, median grain-size (D50) and water content (W0) of the sediment (top 1 cm); (2) the near infrared (NIR) band absorption depth was a key feature in estimating D50 and W0; and (3) the seasonal variation in the total (depth-integrated) MPB biomass was prominent in Chongming Dongtan, Changjiang Estuary. We concluded that the depth-integrated exponential decay algorithm was useful for estimating total MPB biomass and, combined with mappable sediment-surface data, could help map the total MPB biomass in estuarine tidal flats at a large scale.