2002/11/01 by P. D. Martin, D. F. Malley, G. Manning +1 · 2 citations
Environmental Science · Computer Science · Chemistry · #Soil Geostatistics and Mapping #Geochemistry and Geologic Mapping #Spectroscopy and Chemometric Analyses
paper · doi:10.4141/s01-054
openalex publication_date 2002/11/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/06/11
This study explored the use of near-infrared spectroscopy (NIRS) for the rapid analysis of organic C (C org ) and organic N (N org ) in the A horizon of soil within a single field. Soil was sampled throughout a field in Manitoba, Canada to capture soil variability associated with topography. The soil samples were oven-dried and treated with acid to remove carbonates, after which C and N were determined by dry combustion. In this study, portions of the dried soil samples not treated with acid were scanned with a near-infrared scanning spectrophotometer between 1100 and 2500 nm. Correlating the spectral and the chemical analytical data using multiple linear regression or principal component analysis/partial least squares regression gave useful correlations for C org . Over the range of 0–40 mg g -1 C org , NIR-predicted values explained 75–78% of the variance in the chemical results. Results were improved to 80% for calibrations developed for the 0–20 mg g -1 organic C range. Useful results were not obtained for N org although the literature shows that total N in soil is predictable using NIRS. It is likely that the acid treatment altered the composition of the samples in an inconsistent manner such that the chemically analyzed samples and those scanned by NIRS were different from each other in N org concentration or composition. Extrapolation of these C org results to the landscape scale implies that NIRS has potential to be a suitable method for mapping C for the purposes of monitoring C sequestration. Key words: Near-infrared spectroscopy, soil, carbon, nitrogen, topography, soil monitoring