In-situ manipulations of aquatic optical depth and its effect on small unoccupied aerial systems–derived spectral reflectance

Abstract

The collection of spectral data with sensors fixed to various platforms (e.g., satellites, occupied aerial vehicles, and small unoccupied aerial systems (sUAS)) has allowed for the estimation of optically active constituents (OACs) common in surface waters. However, in small, complex, and optically shallow waters where multiple OACs (e.g., chlorophyll-a and total suspended solids) impact the spectral signature, these technologies have experienced significant limitations. Altering the scale at which these examinations are performed on surface waters (e.g., ponds and lakes) to mesocosm systems (37 cm in height and 30 cm in diameter) provides information on the interactions between multiple OACs and insight on the impact aquatic optical depth has on remotely sensed spectra. This field study examines optically shallow and optically deep mesocosm systems simulated in five-gallon buckets to determine the role aquatic optical depth has on developing accurate surface-water quality models. Results demonstrated an accurate representation of OACs in optically deep mesocosms compared with optically shallow mesocosms when assessed with sUAS (i.e., relative percent differences in predicted iron concentrations of −86 and 16 for optically shallow and deep waters, respectively). The interferences observed under these conditions were comparable to literature values when studying optically complex water bodies with hyperspectral data. This study provides a basis for understanding the benefits and limitations of monitoring in-situ water quality via sUAS.

Keywords

• ordinary least squares regression
• mesocosm
• optically active constituents
• remote sensing
• multispectral
• hyperspectral