Characterizing attribute distributions in water sediments by geostatistical downscaling

Area of Science:

• Environmental Science
• Geosciences
• Geostatistics

Background:

• Benthic sediment core sampling provides average attribute values over varying lengths.
• Site characterization and remediation design often require uniform spatial resolution data.
• Spatial autocorrelation of attributes necessitates geostatistical methods for distribution estimation.

Purpose of the Study:

• To optimally infer the spatial distribution of sampled attributes at a uniform resolution from fluvial core sampling data.
• To compare geostatistical downscaling with ordinary kriging for attribute distribution estimation.

Main Methods:

• Formulating geostatistical downscaling as a geostatistical inverse problem.
• Applying the downscaling technique to a hypothetical case study and real-world data (total organic carbon in Passaic River).
• Comparing results with point-to-point ordinary kriging.

Main Results:

• Ordinary kriging inaccurately estimates spatial distribution due to its assumption of uniform data resolution.
• Geostatistical downscaling accurately resolves attribute distribution by accounting for average measurements and fine-resolution data.
• The study demonstrates the effectiveness of geostatistical downscaling for environmental data.

Conclusions:

• Geostatistical downscaling is a superior method for estimating fine-resolution attribute distributions from core sampling data compared to ordinary kriging.
• This technique enhances the accuracy of site characterization and risk-based remediation strategies.
• The study highlights the importance of accounting for sampling resolution in geostatistical analyses.