Photos
Photos
Photos
Photos
Central African soil infrared library
Laura Summerauer, Matti Barthel, Marijn Bauters, Benjamin Bukombe, Mario Reichenbach, Pascal Boeckx, Kristof Van Oost, Basile Bazirake Mujinya, Sebastian Doetterl, and Johan Six
Information about soil properties is not only crucial for efficient and sustainable agricultural production, but also to understand nutrient cycles and fluxes. Compared to traditional wet chemistry analyses of soil samples, infrared spectroscopy produces fast and reproducible results which can overcome spatial heterogeneity. Functional groups of organic matter and mineralogy trigger distinct absorbance peaks in the mid-infrared range. Based on the spectra calibrated to chemical reference values, concentrations of substances can be derived.
In developing countries, infrared spectroscopy can serve as an especially promising tool for soil analysis due to its low costs and simple handling that minimizes chemical consumables. To further reduce the necessity for expensive wet chemistry analyses, large-scale infrared libraries need to be developed.
We have collected a large set of more than 1800 soil samples from past projects in the Congo Basin and along the Albertine Rift, the western branch of the East African Rift System. Samples were scanned on a mid-infrared spectrometer and analyzed using a memory-based learning algorithm. Total carbon and nitrogen were accurately predicted for distinct regions of central Africa–a first step to fill the gap of soil data of this large but understudied area. The resultant open-access library is continuously updated and expanded with additional soil properties and new samples.
