Forest-atmosphere exchange of CO2 isotopes over six years at the Niwot Ridge Ameriflux forest
Abstract
Much of what we know about global and regional carbon cycle processes comes from long-term measurements at sites which are intentionally located away from local sources and sinks. Flux towers provide information on comparatively small scales at locations with important local biological CO2 exchange. To address the inherent spatial scaling mismatch between these approaches, studies are needed which focus on comparisons between them. In the alpine tundra of Niwot Ridge (here called the tundra site), Colorado, above timberline, very long records of CO2 mole fraction (NOAA, since 1968) and the carbon isotope composition (δ13C) of CO2 (U. of Colorado, 1990) exist, providing a wealth of information about continental exchange and insight into free troposphere processes. These data are generated from weekly/biweekly flask samples, and hence have relatively low temporal resolution, but high analytical precision (<0.05 ppm and 0.02 permil). CO2 flux measurements have been conducted at the Niwot Ridge AmeriFlux forest since 1998 - this site is located 500 m lower in elevation and ~4 km distant from the tundra site, and allows for a paired-site comparison, with the tundra site acting as the "background" air influencing and influenced by the forest site. In Sept 2005, we initiated an on-going, long-term study of CO2 and δ13C at the forest and tundra sites. New measurements are of two types: 1) high-precision (< 0.05 ppm) CO2 at the tundra site, and 2) high-precision CO2 (< 0.05 ppm) and δ13C (~0.2 permil) at the Niwot Ridge Ameriflux forest. Both data sets involve multiple measurement heights and sampling frequency is on the order of minutes. Comparisons of the high-frequency data with the flask data at the tundra site show important spatial and temporal differences. Here we show that 1) the standard Keeling plot approach is ineffective at determining the isotopic composition of the local surface-atmosphere net CO2 exchange, particularly in winter, and 2) there are both regional and local carbon cycle signals available in these records. Continued long-term observations that represent processes on a variety of spatial and temporal scales are needed to constrain our understanding of the carbon cycle of the continental United States.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.B44A..07B
- Keywords:
-
- 0315 ATMOSPHERIC COMPOSITION AND STRUCTURE / Biosphere/atmosphere interactions;
- 0426 BIOGEOSCIENCES / Biosphere/atmosphere interactions;
- 0428 BIOGEOSCIENCES / Carbon cycling;
- 0438 BIOGEOSCIENCES / Diel;
- seasonal;
- and annual cycles