File(s) not publicly available
The potential of hyperspectral images and partial least square regression for predicting total carbon, total nitrogen and their isotope composition in forest litterfall samples
journal contributionposted on 2019-07-24, 00:00 authored by I Tahmasbian, Z Xu, K Abdullah, J Zhou, R Esmaeilani, TTN Nguyen, Shahla Hosseini BaiShahla Hosseini Bai
Purpose: The main objective of this study was to examine the potential of using hyperspectral image analysis for prediction of total carbon (TC), total nitrogen (TN) and their isotope composition (δ13C and δ15N) in forest leaf litterfall samples. Materials and methods: Hyperspectral images were captured from ground litterfall samples of a natural forest in the spectral range of 400–1700 nm. A partial least-square regression model (PLSR) was used to correlate the relative reflectance spectra with TC, TN, δ13C and δ15N in the litterfall samples. The most important wavelengths were selected using β coefficient, and the final models were developed using the most important wavelengths. The models were, then, tested using an external validation set. Results and discussion: The results showed that the data of TC and δ13C could not be fitted to the PLSR model, possibly due to small variations observed in the TC and δ13C data. The model, however, was fitted well to TN and δ15N. The cross-validation R2cv of the models for TN and δ15N were 0.74 and 0.67 with the RMSEcv of 0.53% and 1.07‰, respectively. The external validation R2ex of the prediction was 0.64 and 0.67, and the RMSEex was 0.53% and 1.19 ‰, for TN and δ15N, respectively. The ratio of performance to deviation (RPD) of the predictions was 1.48 and 1.53, respectively, for TN and δ15N, showing that the models were reliable for the prediction of TN and δ15N in new forest leaf litterfall samples. Conclusions: The PLSR model was not successful in predicting TC and δ13C in forest leaf litterfall samples using hyperspectral data. The predictions of TN and δ15N values in the external litterfall samples were reliable, and PLSR can be used for future prediction. © 2017, Springer-Verlag GmbH Germany.
Number of Pages13
External Author AffiliationsUniversity Technology Malaysia; Griffith University; University of the Sunshine Coast