Tracing catchment fine sediment sources using the new SIFT (SedIment Fingerprinting Tool) open source software

Area of Science:

• Environmental Science
• Hydrology
• Geochemistry

Background:

• Diffuse sediment pollution requires accurate source identification for effective management.
• Current sediment source fingerprinting methods are complex and less accessible to non-specialists.
• The SedIment Fingerprinting Tool (SIFT) software was developed to address these accessibility and complexity issues.

Purpose of the Study:

• To introduce and evaluate the SIFT software for sediment source fingerprinting.
• To assess if SIFT can reduce uncertainty and increase precision in sediment provenance analysis.
• To explore sediment provenance information using multiple model configurations and uncertainty analyses.

Main Methods:

• Development of the SIFT software, automating calculations and guiding data analysis decisions.
• Application of SIFT to an agricultural catchment in the UK.
• Utilizing multiple source group classifications (k-means clustering, a-priori geological groups) and composite fingerprints.
• Employing various uncertainty analysis methods (bi-plots, range tests, tracer variability ratios, virtual mixtures).

Main Results:

• SIFT integrates recent methodological developments for robust sediment fingerprinting.
• Multiple model configurations and uncertainty assessments were applied to explore provenance information comprehensively.
• Despite some model configurations performing poorly, dominant sediment sources were identified, with proximity being a key factor.
• Different model configurations yielded varying results, highlighting the importance of uncertainty assessment.

Conclusions:

• SIFT provides more robust solutions for catchment management compared to previous approaches.
• The software guides users through complex data processing, improving accessibility.
• Applying multiple configurations and uncertainty assessments leads to more reliable sediment provenance identification.