Landsat observations of total suspended solids concentrations in the Pearl River Estuary, China, over the past 36 years.

Information on long-term trends in total suspended solids (TSS) is critical for assessing aquatic ecosystems. However, the long-term patterns of TSS concentration (C) and its latent drivers have not been well investigated. In this study, we developed and validated three semi-analysis algorithms for deriving C using Landsat images. Subsequently, the long-term trends in C in the Pearl River Estuary (PRE) from 1987 to 2022 and the driving factors were clarified. The developed algorithms yielded excellent performance in estimating C, with mean absolute percentage errors <25% and root mean square errors of <13 mg/L. Long-term Landsat observations showed an overall decreasing trend and significant spatiotemporal dynamics of the C in the PRE from 1987 to 2022. The analysis of driving factors suggested that industrial sewage, cropland, forests and grasslands, and built-up land were the four potential driving forces that explained 87.81% of the long-term variation in C. This study not only provides 36-year recorded datasets of C in estuary water, but also offers new insights into the complex mechanisms that regulate C spatiotemporal dynamics for water resource management.