OOS 7-9 - Evaluating the temporal lags and influence of past climate on tree growth in the Southwest: Implications for responses to climate change

Tuesday, August 9, 2016: 10:50 AM
Grand Floridian Blrm E, Ft Lauderdale Convention Center
Drew M. P. Peltier1,2, Jarrett Barber3 and Kiona Ogle1,2,4, (1)Center for Ecosystem Science & Society, Northern Arizona University, Flagstaff, AZ, (2)Department of Biological Sciences, Northern Arizona University, Flagstaff, AZ, (3)Department of Mathematics & Statistics, Northern Arizona University, Flagstaff, AZ, (4)School of Informatics, Computing, and Cyber Systems, Northern Arizona University, Flagstaff, AZ
Background/Question/Methods

Tree growth affects key ecological, biogeochemical, and climatic processes, yet our understanding of its governing factors requires improvement. Dendrochronologists recognize the importance of recent climate (~1 year prior to ring formation), but we rarely consider older (~2-4 years prior to ring formation) climate variables, which are likely to be important, particularly under physiological stress. We conducted a literature review of correlations between tree-ring widths and monthly climate, followed by an analysis of dendrochronologies from the Southwest for 14 species occurring across 374 sites. We applied the Stochastic Antecedent Modeling (SAM) framework to quantify the temporal lags and relative importance of past climate for tree growth during the 60 months (5 years) prior to ring formation. In particular, we focused on growth responses to temperature, precipitation, and their interaction (via PDSI, a drought index). We explored how growth-climate temporal lags and sensitivities varied among species and across sites. We also quantified how climatic sensitivities have changed over time to evaluate potential indirect impacts of climate change on tree growth, as opposed to the direct effects of changing climatic variables.

Results/Conclusions

Based on our literature review, only ~5% of studies consider older climate (more than one year prior to ring formation), suggesting an opportunity to improve our understanding of growth-climate relationships and their forecasts. Based on our analysis, the main effects of antecedent precipitation and antecedent temperature varied among sites and species, but were generally non-significant (~70-74% of site-species). However, these climate variables are important for tree growth through their interaction, as captured by antecedent drought conditions (PDSI). In particular, higher antecedent PDSI (wetter conditions) stimulated growth in 80% of the sites/species (20% were not significantly affected). The SAM results suggest important lags that varied among species, but overall, stand-level tree growth was most strongly influenced by growing season (spring and early summer) PDSI during the year of growth, and secondarily influenced by PDSI in the previous year’s fall and winter. Stand-level tree growth was also influenced by summer temperatures three years prior to ring formation. We also found evidence that the climatic sensitivities of many species have changed, with the most notable changes occurring between 1940-1965. These results emphasize the importance of older climate variables to characterizing growth-climate relationships and their potential to forecast growth under future climate scenarios.