Journal of Near Infrared Spectroscopy - Abstract of Pinus taeda L. wood property calibrations based on variable numbers of near infrared spectra per core and cores per plantation

Laurence R. Schimleck,^a,* Justin A. Tyson,^a P. David Jones,^b Gary F. Peter,^c Richard F. Daniels^a and Alexander Clark III^d
^aWarnell School of Forestry and Natural Resources, University of Georgia, Athens, GA, USA. E-mail: lschimleck@warnell.uga.edu
^bDepartment of Forestry, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA
^cSchool of Forest Resources and Conservation, University of Florida, Gainesville, FL, USA
^dUSDA Forest Service, Southern Research Station, Athens, GA, USA

Near infrared (NIR) spectroscopy provides a rapid, non-destructive method for the estimation of several wood properties of increment cores. NIR spectra are collected from adjacent sections of the same core; however, not all spectra are required for calibration purposes as spectra from the same core are autocorrelated. Previously, we showed that wood property calibrations that included a single spectrum per core were almost as successful when used to predict the wood properties of sections of new cores, as calibrations based on multiple, consecutive spectra per core. However, it is not known, for calibration purposes, how many NIR spectra should be collected per core, nor how many cores are required to represent a plantation. In this study, we demonstrate that it is unnecessary to use NIR spectra from every section of a core for calibration development. One spectrum per core adequately represents it, provided that sections from other cores representing juvenile, mature and the juvenile/mature wood transition are included in the calibration set. Calibration and prediction statistics can be slightly improved by increasing the number of spectra per core from one to between three and five, with the addition of further spectra unnecessary. For the plantations examined in this study, a minimum of seven cores per plantation is recommended. Increasing the number of cores per plantation to ten (the maximum) is unnecessary and the small improvement is not worth the cost.