Spatial-temporal evolution analysis of pollutants in Daitou River watershed based on Sentinel-2 satellite images

There are large amounts of resources to consume, slow efficiency, limited scope, and lack of access to hazardous areas that are characteristic of traditional methods of assessing pollutants in watersheds. In stark contrast, remote sensing technology offers expansive coverage, swift operation, superior timeliness, and cost-effectiveness, effectively compensating for the shortcomings of traditional methods. The severe siltation and persistent malodorous emissions have long been suffered by the Daitou River due to the substantial influx of wastewater from agricultural, residential, and industrial sources in the surrounding regions. To investigate the water quality status within the Daitou River basin and to delineate the temporal and spatial variations in water quality, the study focuses on the Daitou River basin within the Tongan District of Xiamen City. Utilizing Sentinel-2-L2A satellite imagery from the years 2018 to 2022, a five-year assessment of nitrogen and phosphorus pollutant concentrations in the area was conducted. The methodology entailed image pre-processing to extract the morphological evolution of the study area, followed by the development of an inversion model for nitrogen and phosphorus concentrations based on the sensitive spectral bands of nitrogen and phosphorus concentrations. This model facilitated an analysis of the spatial and temporal distribution patterns, the identification of pollutant fluctuations, the examination of pollution sources, and the implementation of ongoing surveillance on pollutant dynamics within the basin. The findings are as follows: (i) The water area boundary of Daitou River Basin can be obtained from 2018 to 2022, and the area of Daitou River Basin will gradually decrease. (ii) According to the constructed model, it can be concluded that the inversion model of nitrogen and phosphorus concentration with one variable function has the best effect, and the determination coefficients R2 are 0.708 and 0.734, respectively. (iii) Over the period from 2018 to 2022, total nitrogen levels in the Daitou Creek exhibited fluctuating trends, with a marked increase from 2021 to 2022, nearly reaching the threshold of Class V surface water standards by 2022. Conversely, total phosphorus concentrations exhibited an overall declining trend, with a relatively stable period from 2018 to 2021, followed by a significant drop from 2021 to 2022, ultimately aligning with the Class V surface water standards in 2022. The study concludes that the Sentinel-2 image-based pollutant monitoring technique can furnish a more comprehensive, prompt, and quantitative dataset for water quality surveillance. Results of this study aids in refining existing water quality monitoring knowledge and methodologies, thereby enhancing the scientific rigor of pollutant surveillance and remediation efforts. It offers a valuable data source and methodological support for ecological environmental governance assessments, providing a foundation for ecological planning and integrated basin management. Furthermore, it serves as a crucial reference for realizing the intelligent monitoring system of river basin and improving the monitoring capabilities.