[Evaluation of total suspended matter based on spectral classification]

Area of Science:

• Environmental Science
• Remote Sensing
• Water Quality Monitoring

Context:

• Total suspended matter (TSM) concentration is a key water quality parameter.
• Traditional TSM retrieval models often fail for Case 2 waters due to complex optical properties.
• Accurate TSM monitoring is crucial for understanding aquatic ecosystems and managing water resources.

Purpose:

• To develop and validate a water classification method using HJ-1 multispectral data for improved TSM retrieval.
• To differentiate Case 2 waters into distinct optical types for tailored modeling.
• To establish accurate retrieval models for suspended sediment concentration (SSC) specific to water types.

Summary:

• A novel optical classification method was applied to HJ-1 multispectral data to categorize Case 2 waters into two types.
• Two distinct retrieval models were developed: an exponential model for Type I water and a linear model for Type II water.
• The study demonstrates that spectral classification significantly enhances the accuracy of TSM concentration retrieval.

Impact:

• Improved accuracy in retrieving suspended sediment concentration in complex Case 2 waters.
• Provides a more robust method for water quality assessment using remote sensing data.
• Offers tailored modeling approaches for different water optical properties, advancing water resource management.