Temporal and spatial variability in regionalscale CO2 mixing ratios as measured during the MidContin

1
Temporal and spatial variability in
regional-scale CO2 mixing ratios as measured
during the Mid-Continental Intensive study
Natasha Miles, Scott Richardson, Ken Davis, and
Eric Crosson American Geophysical Union Annual Me
eting 2008 17 Dec 2008

ring2.psu.edu
2
Outline

• Mid Continental Intensive, Ring 2
• Temporal variability
• Spatial gradients
• Seasonal cycle
• Comparison to NOAA-ESRL CarbonTracker
• June 2008 flood in midwest
• Explaining small and large spatial gradients
  using a regional flux model (forwards)
  

3
NACP Mid-Continental Intensive
The role of Ring 2 in the Mid-Continent Intensive

• Overall goal Seek convergence between top-down
  (tower-based) and bottom-up (inventory-based)
  ecological estimates of the regional flux
  
• Focus on upper Midwest of the U.S.

• Add regional network of 5 communications-tower
  based atmospheric CO2 obs in the midcontinent
  intensive region
  
• April/May 2007 through October 2008
• Plan to oversample the atmosphere in the
  study region for more than a full year
  
• Inversions to produce maps of fluxes of CO2 for
  region (CSU)
  

WLEF
500 km
4
Instrumentation Details Ring 2

• Cavity ring-down spectroscopy (Picarro, Inc.)
• Long-term stability
• Much reduced need for calibration over NDIR
  methods
  
• Both CO2 and H2O are measured
• 30 and 110-140 m AGL
• Quality control
• Two reference gases at each site, sampled daily
• Round-robin tests (Feb and Nov 08)
• Uncertainty

5
Synoptic variability in boundary-layer CO2 mixing
ratios

• Seasonal drawdown
• Differences amongst the sites
• 2007 vs 2008
• Day to day variability

6
Synoptic variability in boundary-layer CO2 mixing
ratios

• Seasonal drawdown
• Differences amongst the sites
• 2007 vs 2008
• Day to day variability

• Difference in daily value from one day to the
  next as large as 10-30 ppm

7
Temporal variability Night Day CO2

• Difference between nighttime and daytime values
  at 120 m AGL can be over 80 ppm for Ring 2
  
• Average magnitude of the diurnal cycle at 122 m
  for July at LEF 10 ppm (1995-1997) (Bakwin et
  al. 1998)
  

8
Temporal variability Night Day CO2
Ring2
LEF

• Difference between nighttime and daytime values
  at 120 m AGL can be over 80 ppm for Ring 2
  
• Average magnitude of the diurnal cycle at 122 m
  for July at LEF 10 ppm (1995-1997) (Bakwin et
  al. 1998)
  

9
Spatial gradient magnitude (daytime) Growing
seasons 2007-08

• Majority
• But in 6 of cases, the spatial gradient is
  between 0.04 and 0.06 ppm/km (Daytime!)
  

of site-days

• Largest difference amongst the sites for each
  daily value
  

• Seasonal pattern
• Differences as large as 40 - 50 ppm between Ring
  2 sites! Daytime!
  
• Significant day-to-day variability

10
Seasonal cycle

• Strong coherent seasonal cycle across stations
• West Branch (wbi) differs significantly from 2007
  to 2008
  
• Large variance in seasonal drawdown, despite
  being separated by, at most, 550 km. (mm, ce,
  lef) vs (kw, rl, wbi)
  

mlo
11
Seasonal cycle

• Strong coherent seasonal cycle across stations
• West Branch (wbi) differs significantly from 2007
  to 2008
  
• Large variance in seasonal drawdown, despite
  being separated by, at most, 550 km. (mm, ce,
  lef) vs (kw, rl, wbi)
  

mlo
12
Seasonal cycle

• Strong coherent seasonal cycle across stations
• West Branch (wbi) differs significantly from 2007
  to 2008
  
• Large variance in seasonal drawdown, despite
  being separated by, at most, 550 km. (mm, ce,
  lef) vs (kw, rl, wbi)
  

mlo
13
Dominant vegetation map
Corn for Grain 2007Yield per Harvested Acre by
County
Courtesy of K. Corbin
14
NOAA-ESRL Carbon Tracker
Ring2 sites not included as input for 2007
http//carbontracker.noaa.gov
15
2007
14-day smoother applied to CT output
mid-afternoon values only (1930 GMT)
Overall drawdown in CT2008 is too weak, but some
features of modeled variability are consistent
with obs, e.g., there is a lot of variability
and MM has less drawdown than WBI, RL and KW in
both model and obs.
A. Andrews
16
Flooding in the Midwest June 2008
17
Seasonal cycle

• Strong coherent seasonal cycle across stations
• West Branch (wbi) differs significantly from 2007
  to 2008
  
• Large variance in seasonal drawdown, despite
  being separated by, at most, 550 km (mm, ce, lef)
  vs (kw, rl, wbi)
  

18
Delay in seasonal drawdown

• 2008 growing season is uniformly delayed by about
  one month, compared to 2007
  
• Effect of June 2008 flood?
• Recovery increased uptake later in the growing
  season
  

2007
2008
2007 solid 2008 dashed
19
Regional flux model forward results

• SiB3-RAMS coupled with crop phenology model for
  corn and soybean
  
• Addition of crops greatly improved models
  agreement with obs
  
• Gradients of daily minimum value obs vs model
• Simulated CO2 at 120 m above ground level on 16
  July 2007, illustrating advective controls on CO2
  gradients observed across the Ring2 towers

B51A-0364 Corbin et al. Effects of Agricultural
Production on Regional Variations of Atmospheric
CO2 Concentrations
B51A-0355 Lokupitiya et al. Evaluation of the p
erformance of SiBCrop model in predicting carbon
fluxes and crop yields in the croplands of the US
mid continental region
20
Summary

• Temporal variability
• diurnal cycle magnitude large compared to LEF
• 2008 flood in the Midwest US
• Spatial gradients large difference between
  sites despite relatively small site separations
  
• Seasonal cycle
• difference amongst sites, some with very large
  seasonal drawdown
  
• Comparisons to NOAA-ESRL CarbonTracker
• Weak overall drawdown, but good variability