Synoptic CO2 Variability by QCT277


									 Sources of Synoptic CO2
Variability in North America
                 Nick Parazoo
            Atmospheric Science
           Colorado State University
            ChEAS, June 5, 2006

Acknowledgments: Scott Denning, Ken Davis, Arlyn Andrews,
     Doug Worthy, Larry Flanagan, Sebastien Biraud
•   Motivation
•   Data
•   Frontal Identification
•   Preliminary Results
•   Future Research
• Inversion calculations are constrained by
  atmospheric CO2 measurements, which are
  becoming more continental and high
• Continental measurements contain much
  information regarding regional scale
• To take full advantage of the measurements
  requires an understanding of processes
  driving the high-frequency variability
Factors Influencing Continental Surface
CO2 Concentrations (Geels et al., 2004)

• Regional Surface CO2 fluxes from
  biosphere and anthropogenic sources
• Near surface mixing due to BPL growth
  controls signal by changing volume over
  which CO2 flux is spread
• Synoptic meteorological systems are
  responsible for vertical mixing and
  horizontal mixing
Scales of Variability

          • Study focuses the coupling
            of atmospheric dynamics
            and surface fluxes during
            non fair-weather conditions
       Synoptic CO2 Patterns
• Wang 2005 studied
  the effect of frontal
  passage on the CO2
  signal using obs and a
  mesoscale model
• Based on frontal
  identification from
  surface maps and
  data from 1997-2001,
  he found a prominent
  signal during
  summertime frontal
  passages at WLEF
              Research Goals
• Analyze mixing ratios from additional stations, test
  whether similar/other patterns exist across North
  America, and determine whether seasonality to the
  patterns exist
   – Seasonality may be explained in part by mixing of
     horizontal [CO2] gradients across the continent which
     occurs during air mass replacement
   – Szabo et al 2006 studied concentration footprints using
     lagrangian trajectory model (HYSPLIT4)
• Determine contribution of physical and biological
  processes to signal using either coupled
  mesoscale land-surface model or global transport
    Seasonality of N-S gradient
• Gradients are
  controlled by
  differences in
  biospheric flux across
  the continent and
  change sign:
  increases with
  latitude in winter and
  decreases with
  latitude during
  growing season
         Additional Mixing Ratio Data
•   Well calibrated mixing ratio data are
    available at the following stations for
    most of 2003 and 2004:
    1)   WLEF (1997-05, 45.95N, -90.27W),
         Ken Davis, 30-396m, 1hr
    2)   ARM (2002-04, 36.61N, -97.49W),
         Sebastien Biraud, 4, 25, 60m, 1hr
    3)   AMT (2003-05, 45.03N, -68.68W),
         Arlyn Andrews, 12, 107m, 1hr
    4)   Fraserdale (2003-04, 49.88N, -
         81.57W), Doug Worthy, 40m, 1hr
    5)   BERMS (2003-04, 53.98N, -105.12W),
         Doug Worthy, 30m, 1hr
    6)   Western Peatland (2003-05, 54.95N, -
         112.47W), Larry Flanagan, 9m, 30min
    7)   KWKT soon hopefully!
           Site Characteristics
• Represent four different regions in North America
   –   Boreal Canada (BERMS/WPL)
   –   Northeast (AMT)
   –   Upper Midwest (WLEF/FRS)
   –   Southern Great Plains (KWKT/ARM)
• Each region is influenced by different air masses
  and seasonal circulation patterns
   – For example, Higuchi et al 2003 note that Fraserdale
     has four main different seasonal circulation patterns
     and is influenced by three distinct air masses due to
     the location of the Arctic front
     Frontal Locator Function
• Wang identified fronts using combination of surface
  maps and wind observations
• Objective frontal analysis more efficient
• Renard and Clarke define a “frontal locator
  function” based on their definition of a front as a
  “warm-air boundary of a synoptic-scale baroclinic
  zone of distinct thermal gradient,” and later note
  that “frontal-zone boundaries are considered as
  quasi first-order thermal and moisture
        Frontal Locator Function
               
    GG  
•    The time at which magnitude of
     gradient of theta changes the most
     rapidly defines the trough (minimum
     GG) and ridge (maximum GG)
•    McCann and Whistler 2001 suggest
     ‘unnormalized’ version for use with
     any measure of density

    GG   g 
•    Designed for gridded data, try
     applying to time series
   Frontal Locator Using GEOS4

• 1.25x1 deg, 3-hourly,
• Choose nearest grid
  point to station
• Basically matches
  synoptic variability at
• For frontal locator,
  remove diurnal cycle
  using Butterworth filter
         Frontal Locator Test
• Reproduced composite
  CO2 pattern found by
  Wang at WLEF
    More Plots (Winter, LEF)

•   Peak prior to frontal passage. Weather has less impact on
    ecosystem response than in summer case since ecosystem is
    dormant. Time scale of ~ 1 day for signal to mix out.
       More Plots (Fall, LEF)

•   Peak prior to frontal passage. Pattern is similar to winter case
    except stronger amplitude. Weather has more impact on
    ecosystem response than in winter case.
                Future Work
•    Apply frontal locator function to other
     stations and look for patterns
•    Set up modeling experiment to give physical
     explanation to patterns
•    2 options:
    1. SiB-RAMS. Wang reproduced most of signal at
       WLEF for summertime cold front using constant
       background CO2 at boundaries.
    2. SiB-PCTM. Lokupitiya reproduced much of
       synoptic signal at WLEF.
• SiB-PCTM (blue) and
  WLEF-396m (red)
  from Mar23 - Oct17,
• Has difficulties in
  early spring but
  captures variability in
  the summer and early
• Geels, C et al. Investigating the sources of synoptic variability in
  atmospheric CO2 measurements over the Northern Hemisphere
  continents: a regional model study. Tellus B. 2004, 35-50.
• McCann, D. Whistler, J. Problems and solutions for drawing
  fronts objectively. 2001. Meteorol. Appl, 8, 195-203.
• Renard, R. Clarke, L. Experiments in numerical objective frontal
  analysis. Monthly Weather Review. 1965, 547-56.
• Szabo, T. Barcza, Z. Haszpra, L. Aalto, T. Variability in
  Atmospheric CO2 mixing ratio reflected by tall tower
  measurements. 2006. Poster.
• Wang, Jih_Wang. Observations and simulations of synoptic,
  regional, and local variations in atmospheric CO2. Thesis,
Synoptic Variability By Site

To top