Oxygen data assimilation for estimating micro-organism communities' parameters in river systems.

Area of Science:

• Environmental Science
• Hydrology
• Biogeochemistry

Background:

• High-frequency water quality data is crucial for managing urban river systems.
• Data assimilation is gaining traction in hydrology but is underutilized for water quality modeling.

Purpose of the Study:

• Implement a particle filtering algorithm for data assimilation in a hydro-biogeochemical model.
• Assimilate high-frequency dissolved oxygen data to estimate river metabolism parameters.
• Apply the method to the Seine River system.

Main Methods:

• Utilized a community-centered hydro-biogeochemical model.
• Integrated a particle filtering algorithm to assimilate dissolved oxygen data.
• Focused on twelve key parameters identified through sensitivity analysis, including physical and microbial factors.

Main Results:

• Successfully assimilated virtual dissolved oxygen data, matching reference concentrations.
• Parameter identification was influenced by hydrological, trophic, and thermal river conditions.
• Effectively retrieved physical, bacterial, and phytoplanktonic parameters, distinguishing two algal blooms.

Conclusions:

• The particle filtering approach is effective for water quality modeling and parameter estimation in rivers.
• River thermal state significantly impacts parameter identification accuracy.
• The method allows for detailed analysis of phytoplankton dynamics, including circadian cycles.