HIGH-FREQUENCY OBSERVATIONS OF RIVERINE INORGANIC CARBON VARIABILITY USING IN SITU AUTONOMOUS SENSORS

• Main Author: Shangguan, Qipei
• Format: Doctoral or Postdoctoral Thesis
• Language: unknown
• Published: University of Montana 2023
• Subjects: carbon; metabolism; model; river; sensor; sami
• Online Access: https://scholarworks.umt.edu/etd/12207; https://scholarworks.umt.edu/context/etd/article/13328/viewcontent/Shangguan_umt_0136D_10836.pdf

Rivers and streams actively metabolize, emit, transport and store carbon of various forms received from terrestrial environments. Our current understanding of how rivers and streams contribute to the global carbon cycle remains limited, however, in part due to high temporal variability and complexity of inorganic carbon chemistry. In situ autonomous chemical sensors provide powerful tools to observe changes in rivers and streams and enable development of CO2-specific models to quantify riverine processes. In many aspects, our ability to utilize high-frequency sensor measurements remains limited. This dissertation aims to improve our understanding of inorganic carbon cycling through sensor technology and numerical model development. Following a brief introduction, Chapter 2 to Chapter 5 address four different research questions covering topics of sensor design and model applications. Chapter 6 summarizes research findings and unresolved questions for future studies. Specifically, Chapter 2 describes the operation, calibration, and performance of an in situ alkalinity sensor, named the Submersible Autonomous Moored Instrument for Alkalinity (SAMI-alk). Co-deployment of three sensors obtained high-resolution alkalinity time series and revealed critical factors, such as river turbidity, that affect sensor performance. Diel cycles of alkalinity originated from evapotranspiration in the Clark Fork River. Chapter 3 reviews the importance of carbonate buffering in freshwater carbon cycle studies by providing numerical simulations and example sensor time-series. Although buffering has been extensively studied in marine science, it is not always properly integrated into freshwater. Chapter 4 uses novel numerical models to infer river metabolism from dissolved inorganic carbon data, which is contrasted with commonly used dissolved oxygen. The new model was tested at three reaches in the Clark Fork River with single station and two station approaches. The difference between oxygen and carbon metabolism demonstrates the ...