Assessment of resistance to cavitation in Cordilleran cypress using near-infrared spectroscopy

SERGENT AS; CHARPENTIER JP; DALLA SALDA G; FERNÁNDEZ ME; ROZENBERG P; MARTINEZ-MEIER A

Bariloche

Conferencia; International Conference: Adapting forest ecosystems and wood products to biotic and abiotic stress; 2019

LIA Forestia - INTA (Argentina) e INRA (Francia)

Characterizingtrees drought resistance is an important aim in the worldwide context ofincreasing forest decline and dieback. On the other hand, wood demand isincreasing all around the world, urging the need for domestication of nativespecies and selection of the most resistant individuals. So far, relatively fewstudies have focused on tradeoffs between tree functional traits, such asphysiological and wood quality traits; nevertheless, evaluations of both arecrucial to move forestry towards a better adaptation and a most valorization oftheir natural resources. Near infrared spectroscopy (NIRS) is now recognized asan efficient tool to predict wood quality but it is still poorly used topredict wood functional characteristics. In this study, we tested NIRS as ahigh-throughput and non-destructive method for quantifying both resistance tocavitation and wood density in Austrocedrus chilensis.Partial least squares regression (PLS-R) models with cross validation were usedto establish relationships between the NIR spectra and wood traits. Theaccuracy of the models strongly depended on the traits being used forcalibration, with a correct accuracy for P12 and P50 and the highest accuracyfor wood density. This first approach of a NIRS-based prediction of resistanceto cavitation can still be regarded as successful. Thus, the method has highpotential to be put into practice as a rapid, reliable, and non-destructivemethod to determine, in a unique data acquisition, both resistance tocavitation and wood density.