Assessing soil constituents and labile soil organic carbon by mid-infrared photoacoustic spectroscopy

摘要

Recent progress in microphone sensitivity has dramatically increased the performance of Fourier transform mid-infrared photoacoustic spectroscopy (FTIR-PAS). This technique offers benefits over reflectance spectroscopy techniques because the level of sample reflectance has little effect on the PAS signal. This also means that it should be advantageous for soil analysis because of its highly opaque nature. However, only a limited number of studies have so far applied FTIR-PAS to soil characterization and investigation is still required into its potential to determine soil organic carbon (SOC) degradability. The objective of this study was to assess the potential of FTIR-PAS for the characterisation of the labile fraction of SOC and more classical soil parameters, such as carbon and clay content, for a range of 36 soils collected from various field experiments in Denmark. Partial least squares (PLS) regression was used to correlate the collected FTIR-PAS spectra with the proportion of soil organic carbon mineralised after 238 days of incubation at 15 degrees C and pF 2 (C-238d) taken as an indicator of the labile fraction of SOC. Results showed that it is possible to predict total organic carbon content, total nitrogen content and the labile fraction of SOC using FTIR-PAS with an accuracy similar to or better than near infrared (NIR) spectroscopy. FTIR-PAS offered the advantage over NIR of allowing identification of chemical compounds that correlated positively or negatively with the labile fraction of SOC. Spectral bands corresponding to aliphatic, methyl, amide III and polysaccharides compounds were positively correlated with C-238d, whereas bands corresponding to aromatics, amines, amides II and carboxylic acids were negatively correlated with C-238d. In conclusion, FTIR-PAS has proved to be a powerful tool for characterising soil composition and its labile SOC fraction, offering several benefits over reflectance spectroscopy techniques