Simultaneous Multi-Event Kriging for Groundwater Trend Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for mapping groundwater movement near managed aquifer recharge facilities and groundwater pump-and-treat systems are limited by their inability to effectively handle spatio-temporal data sets, leading to inconsistent trend coefficient estimates and inaccurate hydraulic capture zone assessments due to sequential processing of events.
Innovation Solution
The development of simultaneous multi-event universal kriging (MEUK) method, which constructs a block-diagonal kriging matrix to condition trend coefficients across multiple events, allowing for physically-based interpolation of groundwater pressures and flow patterns over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If sequential processing of events is used to map groundwater movement, then each event can be processed independently, but inconsistent trend coefficient estimates and inaccurate hydraulic capture zone assessments occur
Solution Approach 1:
The patent combines multiple event-specific kriging systems into a single unified simultaneous multi-event kriging framework. By merging the processing of multiple events and conditioning trend coefficients across all events together rather than sequentially, the method achieves consistent trend coefficient estimates while maintaining computational feasibility through the block-diagonal matrix structure.
2Measurement precision
If data acquisition frequency is increased to improve monitoring accuracy, then more detailed groundwater pressure data is obtained, but monitoring costs increase
Solution Approach 1:
The patent performs preliminary conditioning of trend coefficients across multiple events using the simultaneous multi-event kriging framework. By pre-conditioning the data using all available events together before generating maps, the method extracts maximum information from existing data, reducing the need for additional frequent data acquisition and thereby lowering monitoring costs while maintaining accuracy.
3Device complexity
If event-specific kriging systems are used for each time occasion, then computational simplicity is maintained, but sufficient data may not exist for each event to estimate trend coefficients and residual semi-variogram
Solution Approach 1:
The patent merges multiple event-specific kriging systems into a unified simultaneous multi-event kriging framework with a block-diagonal matrix structure. This approach pools data across all events to estimate common trend coefficients, ensuring reliable estimates even when individual events have insufficient data, while maintaining computational efficiency through the structured matrix formulation.
Data Source
AI summary
Systems and methods configured to create contour maps of geospatial variables based on hydrometeorological data associated with the variable are described herein. The systems and methods advantageously use simultaneous multi-event universal kriging for spatio-temporal data exploration, analysis and interpolation with the objective of creating a series of related maps, where each map corresponds to a specific sampling event, but wherein some features exhibit spatial relationships persisting over time. In one particular example, water level maps are prepared using the methods, which has the flexibility to allow the conditioning of trend coefficients based on any arbitrary subsets of sample data, and thereby provides a physically based and deterministic rather than wholly-stochastic basis for depicting hydrometeorological data correlations in space and time. The simultaneous MEUK method borrows strength from events possessing a large number of samples to map events possessing fewer data, and allows for implementing “wheel and axle” data analysis.