Geographic Framework
Statistical calculation predicts the distribution of chemical signatures across a geographic area to verify the origin of natural fibers. Creating a spatial variance model involves the analysis of isotopes or trace elements that vary based on local geology and climate. These models allow for the comparison of a sample from a finished product against the expected values of its claimed source location.
Predictive Logic
Researchers collect reference samples from known forest locations to build a continuous map of isotopic or elemental values. The spatial variance model uses interpolation techniques to estimate values for areas where physical samples have not yet been gathered. This allows the system to assign a probability score to a sourcing claim based on how well the sample data fits the modeled landscape.
High resolution models can distinguish between wood grown in different river basins or at different altitudes within the same country. The model assumes that the environmental factors influencing the fiber composition are stable over the harvest period.
Accuracy Threshold
Reliability of the tool increases as more verified samples are added to the underlying database. A spatial variance model provides a scientific alternative to traditional paper based tracking systems that are vulnerable to document tampering.
