Prediction of Peat Properties from Transmission Mid-Infrared Spectra

Abstract. Better understanding of peatland dynamics requires more data on more peat properties than provided by existing databases. These data needs may be addressed with resource efficient measurement tools, such as models that predict peat properties from mid-infrared spectra (MIRS). High-quality spectral prediction models are already used for mineral soils, but similar developments for peatland-focused research lag behind.

Here, we present transmission-MIRS prediction models for peat which are openly available, easy to use, have basic quality checks for prediction quality, and propagate prediction errors. The models target element contents (C, N, H, O, P, S, K, Ca, Si, Ti), element ratios (C/N, H/C, O/C), isotope values (δ¹³C, δ¹⁵N), physical properties (bulk density, loss on ignition (LOI), macroporosity, non-macroporosity, volume fraction of solids, hydraulic conductivity, specific heat capacity, dry thermal conductivity), thermodynamic properties (Gibbs free energy of formation (ΔG_f⁰)), and nominal oxidation state of carbon (NOSC). They are representative for more diverse peat samples than currently existing peat-exclusive models while having a competitive predictive accuracy. Relatively accurate predictions can be made for example for many element contents (C, N, O, S, Si, Ca, ΔG_f⁰, O/C, H/C, bulk density, and LOI).

Many of these properties are not predicted by existing high-quality prediction models focusing on mineral soils. For some of the target variables, high-quality prediction models focusing on mineral soils exist. These models may be more accurate, but reported predictive accuracies are not directly comparable due to imbalances in the amount of organic vs mineral soil samples in the training data. We suggest that some soil properties are easier to predict for peat, whereas others are easier to predict for mineral soils, emphasizing that we need new approaches to meaningfully compare prediction errors of spectral models computed on datasets with variable amounts of organic soils. Our tests also indicate that replacing δ¹³C and δ¹⁵N measurements by MIRS models probably is not feasible due to large prediction errors. Future studies should address the lack of open training and validation data for some peat properties (O, H, NOSC, ΔG_f⁰, LOI, H/C, O/C), the lack of mineral-rich peat samples, and improve and standardize model validation and comparison for models trained on data with very different proportions of peat soils. This study is a step to catch up with high quality standards set by models for mineral soils and it provides novel models for several peat properties. By filling data gaps in the Peatland Mid-Infrared Database, we make a step to provide the data required to better understand peatland dynamics.