The aim of this research project is to develop a robust and easy-to-use method for determining and quantifying the peat content of a substrate. Firstly, this method should make it possible to determine with a high degree of certainty whether a substrate contains peat. Secondly, we aim to use this method to quantitatively estimate the peat content of a substrate.
The overarching aim of the project is to be able to automatically identify the presence of peat in substrates using machine learning techniques. To achieve this, several steps were taken:
- Selection of representative amoeba species typically present in moors;
- Establishment of a standardized sample preparation technique;
- Creation of substrate mixes as reference samples;
- Testing of the peat identification method through biomarkers (long-chain alkanes; this method is much more labor-intensive)
- Creation of a reference data repository containing microscope images of the amoeba shells from 16 different moors;
- Training and validation of machine learning algorithms for the automatic detection of amoeba shells (and therefore the presence of peat) within various substrates;
