Δ17O Trends of Collected Atmospheric CO2 Resulting from Seasonal Changes in the Biosphere
Abstract
The greenhouse gas carbon dioxide (CO2) and the carbon cycle as a whole play a critical role in our understanding of global climate change. In order to constrain the carbon budget, the flux of CO2 between major carbon reservoirs, such as the atmosphere and biosphere, should be quantified. The common tracers used to probe the atmospheric and biogeochemical cycles of CO2 are δ13C and δ18O. More recently, the clumped isotopes Δ47 are also being used. The uncommon isotopes of oxygen, such as 17O, are rarely used because of technical challenges. However, it has been argued that the simultaneous utilization of δ17O and δ18O better constrain the fluxes associated with terrestrial processes [1, 2, 3]. Whole air atmospheric samples are collected at UCSD in a 2-liter bulb routinely (about once every week). Using cryogenic techniques, CO2 is separated from the whole air samples, totally dried, then quantified and measured in a mass spectrometer for δ13C and δ18O. Adopting the method developed by Mahata et al., the CO2 sample is equilibrated with an equal amount of ultra high purity oxygen in the presence of platinum at 700 °C in a quartz reactor for two hours [2]. Thereafter, O2 is separated from the CO2 and δ17O and δ18O of O2 are measured. The isotopic composition of the initial unreacted O2 is also measured for each sample, allowing the δ17O and Δ17O (= δ18O - 0.516 × δ17O) values to be calculated via a projection method. Initial test runs show a reproducibility of less than 0.05‰ (1-σ standard deviation). After ten months of data collection, we find a seasonal trend in Δ17O by applying a moving average to the data. The Δ17O values average 0.2‰ during the summer and fall, but depreciate to about -0.3‰ during the winter and spring. This depreciation may be due to San Diego's more frequent rainfall during the winter, causing an increase in both plant life and CO2 turnover. We further analyze the data by applying a Fourier transform to the Δ17O values and several meteorological parameters (e.g. pressure and relative humidity). We find similar peaks at numerous frequencies, suggesting a correlation between these datasets. Further measurements are ongoing, enabling a better understanding of these results. References: [1] Barkan and Luz (2012) RCMS, 26: 2733 [2] Mahata et al. (2016) RCMS, 30: 119 [3] Thiemens et al. (2014) JGR, 119: 6221.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2016
- Bibcode:
- 2016AGUFM.B53B0531K
- Keywords:
-
- 3322 Land/atmosphere interactions;
- ATMOSPHERIC PROCESSESDE: 0414 Biogeochemical cycles;
- processes;
- and modeling;
- BIOGEOSCIENCESDE: 0426 Biosphere/atmosphere interactions;
- BIOGEOSCIENCESDE: 0428 Carbon cycling;
- BIOGEOSCIENCES