Estimation of suspended particulate matter in turbid coastal waters: application to hyperspectral satellite imagery

Abstract

An empirical algorithm is proposed to estimate suspended particulate matter (SPM) ranging from 0.675 to 25.7 mg L in the turbid Pearl River estuary (PRE). Comparisons between model predicted and in situ measured SPM resulted in Rs of 0.97 and 0.88 and mean absolute percentage errors (MAPEs) of 23.96% and 29.69% by using the calibration and validation data sets, respectively. The developed algorithm demonstrated the highest accuracy when compared with existing ones for turbid coastal waters. The diurnal dynamics of SPM was revealed by applying the proposed algorithm to reflectance data collected by a moored buoy in the PRE. The established algorithm was implemented to Hyperspectral Imager for the Coastal Ocean (HICO) data and the distribution pattern of SPM in the PRE was elucidated. Validation of HICO-derived reflectance data by using concurrent MODIS/Aqua data as a benchmark indicated their reliability. Factors influencing variability of SPM in the PRE were analyzed, which implicated the combined effects of wind, tide, rainfall, and circulation as the cause.

Assessing the Effects of the Hong Kong‐Zhuhai‐Macau Bridge on the Total Suspended Solids in the Pearl River Estuary Based on Landsat Time Series

“…It was estimated that about 75.29 Mt of sediment was transported to the South China Sea between 1955 and 2005 through the PRE (Dai et al, 2007). As observed in previous studies, TSS in the PRE covers a large range from <1 to 140 mg L −1 (Liu et al, 2009; Wang et al, 2018; Zhao et al, 2018). It has noticeable spatial patterns that are characterized by high values close to the mouth of outlets and low values on the seaward side.…”

“…Understanding factors causing this trend behind the scene is crucial for better interpreting the transport of sediment in the PRE, the remediation of local marine ecosystems related to potential damage caused by contaminated sediment, and the protection of marine protected areas. Different factors affect TSS in the PRE, such as soil erosion (Dai et al, 2009; Liu et al, 2018), meteorological and hydrological conditions (Chen et al, 2017; Zhao et al, 2018), dam constructions (Dai et al, 2008; Wu et al, 2012), and sand mining (Hu et al, 2010). These effects have already been comprehensively deciphered in previous studies.…”

“…With respect to the turbid Pearl River estuary, a number of remote sensing related studies for TSS focused on algorithm development and assessment (Dai et al, 2008; Fang et al, 2010; Wu et al, 2013; Xing et al, 2013; Ye et al, 2014; Zhao et al, 2018). Only a few studies concentrated on long time series of TSS based on remote sensing data (Wang et al, 2018; Zhan et al, 2019).…”

Satellite‐Observed Decreases in Water Turbidity in the Pearl River Estuary: Potential Linkage With Sea‐Level Rise

“…All the water turbidity data were spatially binned with a 7  7 pixel window averaging scheme to remove potential speckle noise from the correlation analysis. The single-point sea-level, tidal range, sediment load, and wind speed data were used and assumed to represent the whole study area, as many previous studies have done (e.g., Feng et al, 2014;Li et al, 2019;Zhan et al, 2019;Zhao, Cao, Xu, Ai, et al, 2018;Zhao, Cao, Xu, Ye, et al, 2018). Based on the annual mean series as well as the multi-year seasonal mean series, the coefficients of determination (R 2 ) between turbidity and the above factors were calculated and subsequently plotted as heat maps, to determine the driving factors of long-term turbidity changes; the significantly correlated pixels are highlighted with dots.…”

Evaluation of the Effectiveness of Multiple Machine Learning Methods in Remote Sensing Quantitative Retrieval of Suspended Matter Concentrations: A Case Study of Nansi Lake in North China

“…TSM is one of the important parameters of water quality, which directly affects the transparency and turbidity of the water body [1] and then affects the growth of aquatic organisms and the primary productivity of the water body [2]. Traditional water quality monitoring mainly adopts the method of fixed section and fixed point for sampling analysis, which not only is costly and greatly affected by external conditions but also has defects in large-scale realtime detection.…”