Ankur Desai - Chasing carbon dioxide: The new sport of the 21st century
Significant questions exist on how land and atmosphere interact by exchanging mass and energy, especially in complex terrain. Answering these questions can help improve prediction of the state of the land surface and evolution of weather and climate. A particular interest of mine, as a carbon cycle biogeochemist with a background in boundary layer- and micro- meteorology, is understanding the exchange of carbon dioxide between forests and the atmosphere.
Carbon dioxide (CO2) is an atmospheric trace gas responsible for much of the greenhouse effect that keeps our planet warm. Continued fossil fuel emissions of CO2 is implicated in observed and predicted future global warming that is warming our planet above that caused by the natural cycle of greenhouse gases. However, the fate of emitted CO2 is strongly linked to how forests absorb CO2. The land surface today absorbs nearly half of the emissions of CO2 that we emit and is responsible for most of the observed interannual variability in atmospheric CO2. However, the variability of this effect in space and time is hard to estimate and even harder to predict in face of future climate change and land management.
We use a variety of techniques on the ground, by aircraft, and by orbiting satellites to estimate this variability. This research spans many disciplines and involves foresters, ecologists, hydrologists, meteorologists, and chemists. Much of my research involves working with these diverse groups of scientists and merging observed data and ecosystem models to answer questions on how terrestrial ecosystems absorb and release CO2.
One of the research projects I worked on this year at NCAR was the Airborne Carbon in the Mountains Experiment 2007 (ACME07). One of the goals of ACME07 was to understand carbon exchange in the central Rocky Mountains of the USA. Forests in mountainous regions are poorly studied due to their inaccessibility. Previous assumptions that these regions are not significant players in regional and global carbon budgets have been shown to be incorrect, as most remaining intact forests in the world are mostly found in complex terrain. Changes in frequencies of forest fires, major pest outbreaks, and logging all threaten to alter this landscape and future CO2 exchange.
To conduct this research, we instrumented the NSF/University of Wyoming King Air turboprop airplane for high-precision observations of CO2, carbon monoxide (CO) and oxygen (O2) (Fig. 1).
Figure 1. A view of the instrumented University of Wyoming/NSF King Air research airplane, based in Laramie, WY.
Flight days from March-September were selected to sample a representative number of days across the known seasonal cycle of carbon exchange. On each flight day, advanced high resolution weather forecasting models (including the NCAR Weather Research and Forecasting (WRF) model) were analyzed and fed to a passive particle model run backward in time. Model output gave us a picture of the path of air parcels as they transit due to winds from one region to a selected receptor point (Fig. 2).
Figure 2. Output from the particle model ensemble showing location of particles for given receptor points (noted by colors) at 4 hours and 24 hours prior to 20Z on 8/9/07. Source regions increase in area the further back in time one goes.
As the parcels transit over the mountains, they pick up and lose CO2 due to forest photosynthesis and respiration. Since most weather models do not capture all kinds of flows in mountains, an ensemble of models were used to capture all possible locations of the air parcels to improve future post-processing of the data to back out land-atmosphere CO2 flux. The aircraft then chased this CO2 across space and time over a day or two (Fig. 3)
Figure 3. Morning (purple) and afternoon (pink) flight pattern for a June flight overlayed on Google Earth terrain and cloud imagery. Note the spiral atmospheric profile patterns and airport low passes. Sampling air at low altitudes (50 m above ground) and high altitudes (7000 m above sea level) is required to sample the complexity of flows in mountains. Figure courtesy of Steve Aulenbach, NCAR.
by flying across the landscape (Fig. 4).
Figure 4. View of the Fraser Experimental Forest St. Louis Creek valley near Winter Park, CO from the King Air on low approach. The brown areas reflect regions of trees dead from bark beetle.
Pairing upwind and downwind CO2 observations (Fig. 5)
Figure 5. Sample variability in CO2 across the St. Louis Creek valley in the morning. A major finding of these flights is the large pools of high CO2 that do not vent until late morning. Figure courtesy of Stephan deWekker, University of Virginia.
from the aircraft along with information of mixing in the atmosphere above the mountains yields information on the role of forest CO2 exchange. CO and O2 observations also help for considering the effects of forest fire (which were numerous this summer) and separating the roles of photosynthesis and respiration. The O2 observations were the first research grade airborne O2 observations ever made.
Now that the observation period of ACME07 has just wound down, the hard work now begins on processing and calibrating the observed data, applying the data to ecosystem models, and analyzing the output to test hypotheses about carbon exchange in montaine forests. The small research group includes geographers, ecologists, atmospheric chemists, biogeochemists, and boundary layer meteorologists both at NCAR and at universities. I'm looking forward to continue my work with this diverse group to move from questions to answers.
ASP Spotlight October 2006
For more ASP spotlights click here http://www.asp.ucar.edu/spotlight/archive.jsp