Based on the national scale soil spectral database nitrogen content inversion

Wang, Qianlong

Permalink

https://hdl.handle.net/2142/55862 Copy

Description

Title

Based on the national scale soil spectral database nitrogen content inversion

Author(s)

Wang, Qianlong

Issue Date

2014-06-17

Keyword(s)

Biology, natural substances

Abstract

Fully mining the valid information in soil spectral library, establishing a strong universal inversion model to predict soil total nitrogen (TN) content is one of the important applications of high spectral direction. Studies using partial least squares regression (PLSR) global model, locally weighted regression (LWR) and fuzzy K-means clustering methods combined with PLSR (FKMC-PLSR). 1661 soil samples were collected from 13 provinces in China, which include Tibet, Xinjiang, Heilongjiang, and Hainan. The samples represent 17 soil groups of the Chinese Soil (Genetic) Classification System, and Zhejiang province 104 paddy soil samples to predict. The results show that, under the national scale PLSR global model for high TN values underestimated the prevalence of samples to be predicted, resulting in low overall predictive accuracy; LWR (R$^{2}$ = 0.76, RPD$_{P2}$ = 2.1), especially FKMC-PLSR (R$^{2}$ = 0.82, RPD$_{P3}$ = 2.4) than the local model PLSR (R$^{2}$ = 0.64, RPD$_{P1}$ = 1.4) global model can more accurately inversion TN content. The results can take advantage of the national scale to establish stability and universal spectral database higher content of soil TN forecasting model to provide the necessary information.

Publisher

International Symposium on Molecular Spectroscopy

Type of Resource

text

Language

en

Permalink

http://hdl.handle.net/2142/55862

DOI

https://doi.org/10.15278/isms.2014.TD13

Copyright and License Information

Copyright 2014 by the authors. Licensed under a Creative Commons Attribution 4.0 International License. http://creativecommons.org/licenses/by/4.0/

Owning Collections