Vertical SOC Estimation Using ML and Process-Based Depth Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for estimating Soil Organic Carbon (SOC) are limited by their inability to provide precise, scalable, and spatially variable assessments, especially in heterogeneous landscapes, and existing machine learning models fail to accurately predict SOC beyond the surface layer due to limited satellite penetration and seasonal variability.
Innovation Solution
A method and system utilizing a domain-aware machine learning model that integrates spectral and temporal embeddings with an attention mechanism, combined with a process-based model, to estimate vertical SOC distribution by incorporating satellite data, management proxies, and soil spectral signatures, and applying a correction factor to align with process-based simulations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional machine learning models are used for SOC estimation, then surface SOC can be predicted, but vertical SOC distribution beyond 10 cm depth cannot be accurately estimated due to limited satellite penetration
Solution Approach 1:
The patent extends the estimation from surface level (2D) to vertical depth dimension (3D) by integrating process-based model outputs that simulate SOC distribution across multiple depth layers, allowing vertical SOC profile reconstruction beyond satellite penetration depth
Solution Approach 2:
The process-based model acts as an intermediary that translates surface SOC estimates and environmental drivers into vertical SOC distribution patterns, bridging the gap between limited satellite penetration depth and deep soil carbon assessment needs
2Adaptability or versatility
If conventional ML models are used, then estimation can be performed, but seasonal and regional variability cannot be captured
Solution Approach 1:
The patent incorporates dynamic environmental drivers including time-varying climate data, land use changes, and management practices into the process-based model, enabling the system to adapt to seasonal and regional variations rather than relying on static relationships
Solution Approach 2:
The system dynamically adjusts model parameters based on regional characteristics and seasonal conditions by integrating time-series environmental data, allowing the same framework to reliably predict SOC across diverse climates and time periods
3Measurement precision
If laboratory-based methods are used for SOC estimation, then precise measurements can be obtained, but large-scale spatial assessment becomes time-consuming and expensive
Solution Approach 1:
The patent uses satellite remote sensing data as a proxy copy of ground truth SOC measurements, combined with process-based modeling to infer vertical SOC distribution across large areas without requiring physical soil sampling at each location
Solution Approach 2:
The integrated model framework serves multiple functions simultaneously: it provides surface SOC estimation, vertical profile reconstruction, temporal change detection, and spatial mapping, replacing multiple separate measurement campaigns with a single scalable system
Data Source
AI summary
This disclosure relates generally to a method and system for dynamic estimation of vertical soil organic carbon (SOC) at a region of interest (ROI). State-of-the-art methods are greatly dependent on satellite data. The satellite measurements are reliable for estimating the surface SOC, but it lacks accuracy when it comes to making estimations at sub-surface layers. The present disclosure addresses these problems through a method of dynamic estimation of vertical SOC by combining a domain-aware machine learning (ML) model and a process-based model. The domain-aware ML model incorporates spectral data, management practice data, soil spectral library (SSL) data, and a correction from the process-based model a corrected surface SOC. The process-based model further estimate depth-specific SOC fractions, along with fractions derived from global soil datasets and the SSL. By combining the domain-aware ML model and the depth-specific SOC fractions vertical SOC is estimated.