@misc{10481/82260,
year = {2023},
month = {4},
url = {https://hdl.handle.net/10481/82260},
abstract = {Near-Infrared spectroscopy (NIR) returns full spectra in the region between 750 and
2500 nm. Although a full spectrum provides extremely informative data, sometimes this
enormous amount of detail is redundant and does not bring any additional information. In
this work, different attribute selection methods for the development of vineyard water
status predictive models are presented. Spectra from grapevine leaves were collected onthe-
go (from a moving vehicle) along nine dates during the 2015 season in a commercial
vineyard using a NIR spectrometer (1200e2100 nm). Contemporarily, the stem water potential
(Jstem) was also measured in the monitored vines. A manual selection, based on
Variable Importance in Projection scores (VIP scores) to choose the spectrum intervals
including the most important wavelengths (interval selection), the locally most important
wavelengths in the spectrum (peak selection), as well as the Interval Partial Least Squares
(IPLS) were tested as attribute selection methods. The results obtained for the estimation of
Jstem using the whole spectrum (R2
P ¼ 0.84, RMSEP ¼ 0.167 MPa) were comparable to those
yielded by the three attribute selection methods: the interval selection method (R2
P ¼ 0.80,
RMSEP ¼ 0.186 MPa), the peak selection method (R2
P ¼ 0.77, RMSEP ¼ 0.201 MPa) and the
IPLS (R2
P ~ 0.62e0.79, RMSEP ~ 0.186e0.252 MPa). The highest simplification was provided
by two IPLS models with three wavelengths and bandwidths of 20 and 4 nm that yielded
R2
P~0.78 and RMSEP~ 0.190 MPa. These results corroborate the suitability of a highly
reduced selection of NIR wavelengths for the prediction of grapevine water status, and its
utility to develop simpler multispectral devices for vineyard water status estimation.
© 2023 The Author(s). Published by Elsevier Ltd on behalf of IAgrE. This is an open access
article under the CC BY-NC-ND license.},
organization = {PID2019-108330RA-I00},
organization = {MCIN/AEI/10.13039/501100011033},
publisher = {Elsevier},
keywords = {Grapevine},
keywords = {Stem water potential},
keywords = {Variable Importance in Projection},
keywords = {Scores},
keywords = {Manual wavelength selection},
keywords = {Interval Partial Least Squares},
title = {NIR attribute selection for the development of vineyard water status predictive models},
doi = {10.1016/j.biosystemseng.2023.04.001},
author = {Marañón, Miguel and Gutiérrez Salcedo, Salvador},
}