HR: 1340h
AN: B33D-1068    [Abstracts]
TI: Spatial and Temporal Variation in Carbon and Oxygen Isotopes in Tree Rings: a Comparison of Riparian and Non-riparian Trees
AU: * Brooks, J
EM: Brooks.ReneeJ@epa.gov
AF: Western Ecology Division U.S. EPA/NHEERL, 200 SW 35th St, Corvallis, OR 97333 United States
AU: Barbour, M M
EM: BarbourM@landcareresearch.co.nz
AF: Maanaki Whenua - Landcare Research, PO Box 69, Gerald Street, Lincoln, 8152 New Zealand
AU: Lee, E
EM: Lee.ehenry@epa.gov
AF: Western Ecology Division U.S. EPA/NHEERL, 200 SW 35th St, Corvallis, OR 97333 United States
AB: Understanding the natural spatial and temporal variability of processes within watersheds will be central to our ability to understand how human-induced changes are altering these processes. Tree rings and the stable isotopes preserved in cellulose provide a long term record of physiological responses to climate variability. By analyzing trees distributed within a watershed, we can add a spatial component to the temporal record in tree rings. The oxygen isotope in tree-ring cellulose contains signals from both source water and evaporated leaf water, but separating these two signals is often difficult. The comparison of riparian trees with upland trees may be one approach for separating these signals since riparian trees will have less variable water source compared to upland trees. We tested this idea using riparian trees from a spring-fed river with a stable source-water signal over time, and compared their δ¹⁸O patterns with those of more upland trees approximately 1/2 mile from the river. 150-year tree-ring chronologies from 5 riparian and 5 upland trees were analyzed for annual growth, Δ¹³C and δ¹⁸O in cellulose. The δ¹⁸O signals from riparian and upland trees were highly correlated with each other, more so than Δ¹³C and growth which were also correlated between the two sites. As expected, δ¹⁸O for both sites was negatively correlated with the average relative humidity during the growing season (May-August) indicating a leaf-water signal. However, the correlation was similar between sites indicating that riparian trees did not have a stronger leaf-water signal compared to upland trees as we might have expected. One reason could be that the upland trees may also have access to deep spring water given the local geology. If source water caused significant variation in annual δ¹⁸O values, we might expect a strong correlation with winter temperature or precipitation, but neither was significant for either site. Only growing season variables which might effect leaf evaporation were significantly correlated with δ¹⁸O values. Interestingly, growth was strongly positively associated with δ¹⁸O, but not correlated with relative humidity or Δ¹³C. The only climate variables significantly correlated with growth were summer temperatures with high summer temperatures associated with decreased growth. Our results indicate that multiple isotopes from trees across a watershed that vary in source-water stability may prove to be a useful technique for interpreting tree-ring signals and watershed changes over time.
DE: 0439 Ecosystems, structure and dynamics (4815)
DE: 0454 Isotopic composition and chemistry (1041, 4870)
DE: 0476 Plant ecology (1851)
SC: Biogeosciences [B]
MN: Fall Meeting 2005