Remote sensing identification of urban water pollution source types using hyperspectral data

Area of Science:

• Environmental Science
• Remote Sensing
• Water Quality Monitoring

Background:

• Urbanization leads to increased water pollution, necessitating effective source identification for treatment.
• Current methods for pinpointing urban water pollution sources are limited.

Purpose of the Study:

• To establish a remote sensing identification method for urban water pollution sources using UAV hyperspectral images.
• To classify different types of urban water pollution sources.

Main Methods:

• Utilized 3D excitation-emission matrix fluorescence spectroscopy and parallel factor analysis to identify seven fluorescent components of dissolved organic matter (DOM).
• Applied hierarchical cluster analysis on fluorescence components and spectral indices to determine four pollution source (PS) types.
• Developed an identification method using water color and optical parameters (e.g., DOM absorption, chlorophyll-a) for UAV hyperspectral data.

Main Results:

• Four distinct pollution source types were identified: domestic sewage, terrestrial input, agricultural and algal, and industrial wastewater.
• The developed method achieved over 70% recognition accuracy for the four PS types.
• Demonstrated the potential of combining DOM fluorescence and water optical properties for PS identification.

Conclusions:

• The study successfully demonstrated a novel remote sensing approach for urban water pollution source identification.
• This method enhances urban water quality management by providing reliable and comprehensive information.
• Expands the application of remote sensing technologies in environmental monitoring.