LAUSR

Normalized water-leaving radiance spectra nLw(λ) at the near-infrared (NIR) from five years of observations (2012–2016) with the Visible Infrared Imaging Radiometer Suite (VIIRS) onboard the Suomi National Polar-orbiting Partnership (SNPP) are used to derive the particle backscattering coefficients bbp(λ) for global highly turbid coastal and inland waters. Based on the fact that the absorption coefficient of sea water aw(λ) is generally much larger than those of the other constituents aiop(λ) at the NIR wavelengths in coastal and inland waters, a NIR-based bbp(λ) algorithm for turbid coastal and inland waters has been developed and used in this study. This algorithm can be safely used for highly turbid waters with nLw(745) and nLw(862) less than ∼6 and ∼4 mW cm−2 μm−1 sr−1, respectively. Seasonal and inter-annual variations of bbp(λ) in China's east coastal region, the Amazon River Estuary, the La Plata River Estuary, the Meghna River Estuary, the Atchafalaya River Estuary, and Lake Taihu are characterized and quantified. The coefficient bbp(λ) can reach over ∼3–4 m−1 in the Amazon River Estuary and China's east coastal region. The Amazon River Estuary is identified as the most turbid region in the global ocean in terms of bbp(λ) magnitude. bbp(λ) spectra in these five highly turbid regions are also seasonal- and regional-dependent. In the highly turbid waters of China's east coastal region and the Amazon River Estuary, bbp(λ) generally increases in wavelength from 410 to 862 nm, while it decreases in the La Plata River Estuary and Atchafalaya River Estuary. This is attributed to the different particle size distributions in these waters. The geophysical implication of the bbp(λ) spectral curvatures for different waters is discussed. To improve global bbp(λ) for both open oceans and coastal turbid waters, a new combined NIR- and Quasi-Analytical Algorithm (QAA)-based bbp(λ) algorithm is proposed and demonstrated.