Rivers and streams play a critical role in connecting major global carbon pools – land, oceans, and atmosphere. The longitudinal flow in rivers and streams allows for the integration and transport of fluvial particulate organic carbon (POC) from various sources and its continual transformation. A continuum-oriented perspective suggests longitudinal variations in POC sources and its dynamics, primarily based on spatial gradients in stream geomorphology and downstream decoupling of the stream channel from adjacent landscapes. However, this prediction has not been comprehensively examined with adequate spatial and temporal resolution. This dissertation aims to develop a higher spatiotemporal resolution approach to enhance the understanding of fluvial POC dynamics from the continuum perspective, with a focus on identification of POC sources and their relative importance changes during storm events when POC inputs and exports are maximized. In the Clear Creek watershed, IA, a well-characterized agricultural watershed, storm POC was collected at high-frequency intervals from three sampling sites from the upper to the lower reaches. Broad-spectrum source-specific biomarkers, obtained from the TMAH-assisted thermochemolysis, and stable carbon isotopes were used as primary source identifiers. Heatmaps with clustering were employed for efficient visualization and interpretation of the complex high-temporal resolution biomarker data. Intra-event POC source shifts and their downstream variations during a storm event were first investigated. A temporal sequence of POC inputs was consistently observed at all stations, from algal and microbial OC (pre-event POC) to terrestrial OC with vascular plant signal (event POC). The event POC was time-resolved into early flush POC dominating in the upper reaches and main POC dominating in the lower reaches. This was attributed to the higher contribution of surface erosion triggered by precipitation in the upper reaches and the enhanced contribution from tributaries and channel/bank erosion in the lower reaches. These spatial variations were generally observed across multiple storm events, with the lower reaches exhibiting less time-resolved biomarkers. However, spatial and temporal variations in POC sources were influenced by various environmental factors, such as spatially heterogeneous precipitation patterns, storm flashiness, antecedent conditions, and seasonal variations in land cover. A collective knowledge of these factors is crucial in understanding event-scale POC dynamics comprehensively. Moreover, an untargeted analysis conducted on a subset of the dataset (collected from the upper region of Clear Creek during a single storm event) revealed the potential for improved resolution in POC sources. Despite limited compound identification, many previously untargeted compounds (i.e., biphenyl-like compounds, odd carbon number fatty acids, saccharides, etc.) were shown to be time-resolved, enhancing the identification of storm POC sources. Incorporating these compounds into the analysis allowed for further time-resolved categorization of the event POC into lignocellulose-rich POC and long-chain fatty acids-rich POC. This dissertation lays the foundation for further higher resolution approaches to advance our understanding of POC dynamics in stream networks.

Creator

• Kim, Jieun

DOI

• https://doi.org/10.21985/n2-mm03-2s75

Subject

• Environmental science
• Biogeochemistry

Language

• en

Alternate Identifier

• http://dissertations.umi.com/northwestern:16775
• etdadmin_upload_1014167

Keyword

• biomarker
• particulate organic carbon
• storm
• stream
• agricultural watershed

Date created

• 2023-01-01

Resource type

• Dissertation

Rights statement

• In Copyright