© 2006 Heron Publishing—Victoria, Canada
Co-occurring species differ in tree-ring δ18O trends
John D. Marshall (1, 2) and Robert A. Monserud (3)
1. Department of Forest Resources, University of Idaho, Moscow, ID 83843, USA / 2. Corresponding author (jdm@uidaho.edu) / 3. Pacific Northwest Research Station, USDA Forest Service, 620 SW Main Street, Suite 400, Portland, OR 97205, USA / Received July 18, 2005; accepted November 15, 2005; published online May 1, 2006
Summary
The stable oxygen isotope ratio (δ18O) of tree-ring cellulose is jointly determined by the δ18O of xylem water, the δ18O of atmospheric water vapor, the humidity of the atmosphere and perhaps by species-specific differences in leaf structure
and function. Atmospheric humidity and the δ18O of water vapor vary seasonally and annually, but if the canopy atmosphere is well mixed, atmospheric characteristics should
be uniform among co-occurring trees. In contrast, xylem water δ18O is determined by the δ18O of water being drawn from the soil, which varies with depth. If co-occurring trees draw water from different soil depths,
this soil-water δ18O signal would be manifest as differences in δ18O among the trees. We examined the variation in tree ring δ18O, over eight decades during the 20th Century, among three species co-occurring in natural forest stands of the northern Rocky
Mountains in the USA. We sampled 10 Douglas-firs (Pseudotsuga menziesii (Mirb.) Franco var. glauca), 10 ponderosa pines (Pinus ponderosa Laws.) and seven western white pines (Pinus monticola Dougl.). As expected, variation in atmospheric conditions was recorded in the δ18O of the cellulose produced in a given year, but observed climatic correlations with δ18O were weak. Significant correlations with June climate data included: daily maximum temperature (r = 0.29), daily minimum temperature (r = –0.25), mean temperature (r = 0.20), mean daily precipitation (r = –0.54), vapor pressure deficit (r = 0.32) and solar radiation (r = 0.44). Lagged effects were observed in Douglas-fir and western white pine. In these species, the δ18O of a given annual ring was correlated with the δ18O of the previous ring. Ponderosa pine showed no significant autocorrelation. Although the species means were correlated among
years (r = 0.67 to 0.76), ponderosa pine was consistently enriched in δ18O relative to the other species; differences were close to 2‰ and they are steadily increasing. Relative to the mean for the
three species, ponderosa pine is becoming steadily more enriched (–1.0‰). In contrast, Douglas-fir is being steadily depleted
and western pine is intermediate, with an enrichment of 0.5‰. Because all trees were exposed to the same atmospheric conditions,
the differences in δ18O observed between species are likely due either to differences in the depth of water extraction or leaf function. If the
former, presumably ponderosa pine has steadily taken up more water from near the soil surface and Douglas-fir has shifted
uptake to a greater depth. If the latter, we suggest the pronounced changes in leaf-water δ18O are a result of changes in leaf structure and function with tree size and age.
Keywords:
cellulose, climate, conifers, Pinus monticola, Pinus ponderosa, Pseudotsuga menziesii, stable oxygen isotope ratio, tree rings.
