Soil Mapping Model Refinement Through Uncertainty-Guided Sampling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing soil mapping techniques face challenges in achieving accurate and efficient determination of sampling locations for measuring soil characteristics, often resulting in inaccurate soil maps, particularly in areas dissimilar to those where quantitative soil data was collected.

Innovation Solution

A method involving training a soil mapping model by estimating uncertainty at various locations, selecting high-uncertainty areas for additional sampling, and refining model parameters based on collected samples to improve accuracy and reduce the number of required samples.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If digital soil mapping techniques using environmental variables are used to predict soil characteristics, then the cost and time required for soil mapping is reduced, but the accuracy of the soil map deteriorates, particularly in areas dissimilar to training areas

Engineering Contradiction:
Improvesoil mapping efficiencyVSAvoidsoil map accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary uncertainty analysis using the initial soil mapping model to identify high-uncertainty areas before collecting additional soil measurements. This preliminary identification of problematic areas allows targeted sampling to be performed in advance of final map generation, improving accuracy where it is most needed while maintaining efficiency across the entire study area

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies different data collection and modeling strategies to different regions based on their specific uncertainty characteristics. High-uncertainty areas receive additional targeted soil measurements and refined modeling, while low-uncertainty areas rely on the initial environmental variable-based predictions, optimizing resource allocation and improving overall map accuracy

Inventive Principle:
Principle #3Local quality

2Measurement precision

If a large number of high-quality quantitative soil measurements are collected across the area of interest, then the accuracy of the soil map is improved, but the time and cost required for data collection increases

Engineering Contradiction:
Improvesoil map accuracyVSAvoiddata collection time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system concentrates measurement efforts in high-uncertainty areas identified by the uncertainty analysis, rather than distributing measurements uniformly across the entire study area. This localized focus ensures that measurements are obtained where they will have the greatest impact on improving map accuracy, reducing the total number of measurements needed while maintaining or improving overall accuracy

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system uses the initial soil mapping model and uncertainty analysis to provide feedback on where additional measurements are most needed. This feedback loop guides the selective collection of additional soil measurements, ensuring that data collection efforts are directed to areas that will most improve model accuracy, thereby reducing unnecessary measurements and saving time

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4314906B1Systems and methods for soil mapping
Publication Date: 2026.03.04 TERRAMERA INC
  • EP4314906B1 patent drawingFigure 1A~1B
  • EP4314906B1 patent drawingFigure 2
  • EP4314906B1 patent drawingFigure 3

AI summary

Systems and methods for training a soil mapping model are provided, as well as systems and methods for generating soil maps with such trained soil mapping models. The soil mapping models predict soil characteristics (e.g. carbon content) across an area of interest. The training method can involve refining the parameters of a soil map model by estimating the soil mapping model's uncertainty at various locations across an area of interest, selecting relatively high- uncertainty areas, identifying sampling locations within those areas for further samples to be collected, and further training the parameters of the soil mapping model based on the newly-collected samples.