Needle cell elongation and maturation timing derived from pine needle cellulose delta18O.

Estimates of the timing of Pinus arizonica Engelm. needle development in 1998 and 1999 were derived from the leaf-cellulose delta18O of weekly growth increments. Significant correlations were noted between time series of local humidity and leaf-cellulose delta18O for needles growing near Tucson, Arizona. Correlations with temperature were also significant, but much lower, suggesting these variations in cellulose delta18O were determined mostly by changes in humidity. The timing of all significant correlations lags the timing of the appearance of the new needle growth, and is interpreted as indicating 16-23 d were required for cell enlargement in 1998 and 13-17 d in 1999. Similarly, properties of the environmental time series, when significantly correlated, are interpreted as indicating the duration of cellulose deposition (7-27 d in 1998, 13-21 d in 1999). Variations in stable-isotope back diffusion (the Péclet effect) and the synthesis of cellulose using stored photosynthate are discussed as explanations for departures from a Craig and Gordon-type model of leaf water delta18O. The Péclet effect, use of stored photosynthate, and variations in the growing-season source-water delta18O, probably confound the development of a high-resolution paleohumidity proxy from subfossil needle cellulose delta18O in this region.