Glacial Tree Physiology: Using Stable Isotopes to Reconstruct Plant Responses to Environmental Change Since the Last Glacial Period

Abstract

Increases in atmospheric [CO2] (CO2 concentration) over the last several hundred years have resulted in a current level of just under 400 ppm and represent novel conditions for modern plants relative to their glacial counterparts. Glacial plants experienced consistent oscillations in [CO2] between 180 and 270 ppm coinciding with glacial-interglacial cycles of the last ~1 million years. Studies of modern plants grown under glacial [CO2] show severe and consistent negative responses in physiology and biomass; however, detailed analysis of glacial plant material remains limited. Investigation of long-term plant responses to changes in atmospheric CO2 levels provides important information on glacial plant physiological patterns as well as ecosystem-level processes such as primary productivity and terrestrial carbon storage. To assess plant responses to low [CO] over geologic time scales, preserved glacial wood material was analyzed and compared to modern trees from the same regions. Glacial Juniperus specimens spanning the last 50,000 years were obtained from the La Brea tar pits in Los Angeles, CA. Glacial Agathis specimens, 50,000+ years old, were obtained from peat bogs in North Island, New Zealand. In both systems, ring width and carbon isotope analysis was performed to compare physiological responses to changes in [CO2] and environmental factors since the last glacial period. Carbon isotopic signatures were used to calculate ci/ca (the ratio of internal CO2 availability to that of the atmosphere) and ci. Oxygen isotope analysis was also performed on Juniperus to analyze responses to anomalous events, specifically El Niño years. Both Juniperus and Agathis showed constant mean ci/ca between the last glacial period and modern times. Glacial mean ci was half the modern ci levels in both species. These results suggest severe carbon limitations in glacial trees, which could have impacted primary productivity and annual growth patterns. Despite having less than half the available carbon, glacial Juniperus and Agathis were able to maintain similar growth patterns to their modern counterparts. We attribute this lack of CO2 fertilization on tree growth to environmental constraints specific to each region, and constraints resulting from adaptations to 10-14 million years of low CO2 conditions. Oxygen isotope analysis was performed on glacial and modern Juniperus to reconstruct El Niño impacts in southern California over the last glacial period using a Bayesian model developed on low-elevation southern California Juniperus. Under less water-limited growing conditions, modern Juniperus from higher elevations do not respond as strongly or as predictably to ENSO-induced changes in temperature and precipitation. This result suggests the same could be true for glacial trees, which could confound proxy-based results in this region. A deeper understanding of the climate-physiology relationship of a species under different environmental conditions is required before a reliable paleo-proxy can be developed. This research advances our understanding of plant responses to glacial conditions. Carbon isotope analysis provides some of the first direct evidence that glacial plants remained near their lower carbon limit throughout the last glacial period. The ring width analysis shows that operating under limiting carbon conditions did not reduce growth in glacial trees, likely due to environmental constraints on growth, and adaptive and evolutionary constraints to utilizing higher [CO2] availability. The oxygen isotope analysis indicates altered physiological strategies under less water-limited growing conditions, which impact the strength of plant responses to anomalous climatic events. Collectively, these results have serious implications for understanding of glacial plant function, estimating ecosystem-scale responses such as primary productivity, and developing paleo-proxies for global atmospheric circulation patterns.