INSTITUTO ARGENTINO DE NIVOLOGIA, GLACIOLOGIA Y CIENCIAS AMBIENTALES
Unidad Ejecutora - UE
Modelling tree ring cellulose <i>δ</i><sup>18</sup>O variations in two temperature-sensitive tree species from North and South America
BÉGIN, CHRISTIAN; SAVARD, MARTINE M.; RISI, CAMILLE; GUIOT, JOËL; NAULIER, MAUD; DAUX, VALÉRIE; LAVERGNE, ALIÉNOR; VILLALBA, RICARDO; BOUCHER, ETIENNE; GENNARETTI, FABIO
Climate of the Past
European Geophysical Union
Año: 2017 vol. 13 p. 1515 - 1526
Oxygen isotopes in tree rings (!18OTR) are widelyused to reconstruct past climates. However, the complexityof climatic and biological processes controlling isotopicfractionation is not yet fully understood. Here, we use theMAIDENiso model to decipher the variability in !18OTR oftwo temperature-sensitive species of relevant palaeoclimatologicalinterest (Picea mariana and Nothofagus pumilio) andgrowing at cold high latitudes in North and South America.In this first modelling study on !18OTR values in both northeasternCanada (53.86! N) and western Argentina (41.10! S),we specifically aim at (1) evaluating the predictive skill ofMAIDENiso to simulate !18OTR values, (2) identifying thephysical processes controlling !18OTR by mechanistic modellingand (3) defining the origin of the temperature signalrecorded in the two species. Although the linear regressionmodels used here to predict daily !18O of precipitation(!18OP) may need to be improved in the future, the resultingdaily !18OP values adequately reproduce observed (fromweather stations) and simulated (by global circulation model)!18OP series. The !18OTR values of the two species are correctlysimulated using the !18OP estimation as MAIDENisoinput, although some offset in mean !18OTR levels is observedfor the South American site. For both species, thevariability in !18OTR series is primarily linked to the effectof temperature on isotopic enrichment of the leaf water. Weshow that MAIDENiso is a powerful tool for investigatingisotopic fractionation processes but that the lack of a denserisotope-enabled monitoring network recording oxygen fractionationin the soil?vegetation?atmosphere compartmentslimits our capacity to decipher the processes at play. Thisstudy proves that the eco-physiological modelling of !18OTRvalues is necessary to interpret the recorded climate signalmore reliably.