Learning Center
Plans & pricing Sign in
Sign Out



									            Annual Pattern of Carbon Exchange and Evapotranspiration
              of an Upland Hardwood Forest in Northern Wisconsin

                   Kenneth J. Davis, Bruce D. Cook, and Weiguo Wang
                               (Brad? Ron? Jud? Paul?)

                              Department of Meteorology
                          The Pennsylvania State University
                                  503 Walker Building
                         University Park, PA 16802-5013 USA

Date of receipt:

Keywords (6):

Corresponding author:

       Kenneth J. Davis
       Department of Meteorology
       The Pennsylvania State University
       503 Walker Building
       University Park, PA 16802-5013 USA
       ph 814-863-8601
       fax 814-865-3663

Running title (45 characters):



                                Materials and Methods

Site description

   Measurements were collected in an upland hardwood forest of the Chequamegon-
Nicolett National Forest in northern Wisconsin (45° 48.47' N, 90° 04.72' W, elevation
515 m). The 60 to 80 year-old stand has a closed canopy approximately 24 m in height,
and consists primarily of sugar maple (Acer saccharum), basswood (Tilia americana),
and green ash (Fraxinus pennsylvanica). The understory consists primarily of suger
maple and ironwood (Ostrya virginiana) saplings, maidenhair (Adiantum pedatum) and
bracken (Pteridium aquilinum) ferns, and blue cohosh (Caulophyllum thalictroides). The
stand occupies about 260 ha, and is homogenous except for a small strip (about 80 m
wide and 500 m long) of red maple (Acer rubrum), black ash (Fraxinus nigra), and
slippery elm (Ulmus rubra) located about 250 m east of the observation tower. The
terrain is relatively flat, ranging from 490 to 520 m, and the slope at the tower base is
about 1% (downward towards the west).

  Logging history?

Eddy covariance measurements

   A Rohn 45G tower was instrumented at 29.6 m above the soil surface for carbon and
energy flux measurements using the eddy covariance method. Fluxes of CO2, latent
heat, and sensible heat were calculated using methods described by Berger et al.
(2001). A three-dimensional sonic anemometer (Campbell Scientific Instruments,
Logan, UT, model CSAT-3) was used to measure wind speeds and sonic virtual
temperature, and an infrared gas analyzer (Li-Cor, Lincoln, NE, model LI-6262) was
used to measure CO2 and H2O vapor mixing ratios. Measurements were collected at a
freqency of 10 Hz, and the LI-6262 was continuous calibrated with precise CO2 storage
measurements and a chilled mirror hygrometer (EdgeTech, Milford, MA, model 200
DewTrak) or relative humidity probe (Cambell Scientific Instruments, model CS500) as
described by Berger et al. (2001). The anemometer and air sample inlet were attached
to a 2 m boom pointed in the predominate wind direction (west). Air was drawn through
a teflon filter (1 m pore size) and about 3 m of teflon PFA tubing (1.56 mm id) before
passing through the LI-6262. Measurements with the LI-6262 were made in absolute
mode by maintaining a constant flow of CO2- and H2O-free N2 gas through the
reference cell.

  Estimated flow rate and pump?

CO2 Storage measurements

  Mean mixing ratio profiles of CO2 were measured at 0.6, 1.5, 3.0, 7.6, 13.7, 21.3, and
29.6 m above the soil surface using a Li-Cor LI-6251 infrared gas analyzer. A similar
system was described by Zhao et al. (1997) and Bakwin et al. (1995) for obtaining high

precision CO2 measurements at unattended sites. Solenoid valves were used to control
flow from each of the levels and each of three compressed gas standards (340, 440,
and 550 ppm CO2 in dry air) through the sample cell of the analyzer. Measurements
were made in differential mode by maintaining a constant flow of a compressed gas
standard containing 440 ppm CO2 in dry air. A back flow pressure regulator (Porter
Instrument, Hatfield, IA, model 9000) was used to equalize the pressure in the sample
and reference cells.
   Precise concentrations of CO2 in working gas standards were determined with a LI-
6251 that was calibrated in differential mode with 340 and 550 ppm CO2 standards
that were prepared by the Climate Monitoring and Diagnotics Laboratory, National
Oceanic and Atmospheric Administration (Kitzis and Zhao, 1999). Air samples were
drawn through a teflon filter (1 m pore size) and variable lengths of Dekabon tubing
(Saint-Gobain Performance Plastic, Wayne, NJ, model 1300, 5.5 mm id) to the LI-6251
at the tower base. Air entering the sample cell was dried in a Nafion drier (Permapure,
Toms River, NJ, model MD-050-72P with a countercurrent of N2 gas) followed by a
chemical desicant (Cl2MgO8). Air was drawn continuously through each tube, and each
height was analyzed for 3 minutes every 21 minutes. Two minutes were required to
flush the system, and actual measurements were collected and averaged during the
final minute. Reference cell gas was passed through the LI-6251 sample cell every 42
minutes to obtain the zero drift, and a sequence of all three standards was measured
every 3 to 4 hours. Standard data were fit with a second order polynomial, and
calibration coefficients were interpolated using a cubic spline function (Research
Systems Inc., 1998) to obtain calibration equations for each 3 minute measurement.

Climate measurements

Soil temperature/moisture profile
CS500 temperature profile
Soil heat flux

Net Ecosystem Exchange calculations

  Turbulent flux calculation
  Change in storage

Data screening and gap filling


Climatic conditions

Energy closure

Evapotranspiration and surface energy partitioning

Ecophysiological relationships

Net Ecosystem Exchange (NEE) of carbon


Ron/Aaron, Art, Tom/Gary/Karla, Jon, Peter, Dana


Bakwin PS, Tans PP, Zhao CL, Ussler W, Quesnell E (1995) Measurements of carbon
dioxide on a very tall tower. Tellus, 47B, 535-549.

Berger BW, Davis KJ, Yi C, Bakwin PS, Zhao CL (2001) Long-term carbon dioxide
fluxes from a very tall tower in a northern forest: Flux measurement methodology.
Journal of Atmospheric and Oceanic Technology (in press).

Kitzis D, Zhao CL (1999) CMDL/Carbon Cycle Greenhouse Gases Group Standards
Preparation and Stability. National Oceanic and Atmospheric Administration Technical
Memorandum ERL-14, Climate Monitoring and Diagnostics Laboratory, Boulder,
Colorado, 14 p.

Research Systems Incorporated (1998) Interactive Data Language Reference Guide,
Version 5.1 Edition, Research Systems Incorporated, Boulder, Colorado.

Zhao CL, Bakwin PS, Tans PP (1997) A design for unattended monitoring of carbon
dioxide on a very tall tower. Journal of Atmospheric and Oceanic Technology, 14,1139-

Table 1. Parameter estimates for filling gaps in observed net ecosystem exchange
(NEE) during 2000.

                        Respiration coefficients§            Photosynthesis coefficients¶
                        ___________________                  ______________________

Month            a0       a1      a2           b0       b1     b2

January                 1.32      0.0344     34.1             0.937        -7.64    1.19

February              14.8        0.267      12.4             1.04       -22.1      1.42

March                   8.77      0.195      14.0             1.28       -21.4      1.70

April                   5.28      0.111      19.4             1.17       -17.7      1.53

May                     7.04      0.143      16.0            13.5       192         0.704

June                    7.32      0.138      14.8            37.4       232         4.86

July                    7.04      0.127      15.3            46.2       413         2.36

August                  6.78      -0.145     19.6            48.4       381         6.06

September               6.90      0.131      15.5            22.7       437        -1.56

October                 6.12      0.132      16.2             1.15         -7.08    0.768

November                8.18      0.195      14.6             2.50         -5.52    2.51

December         3.20    0.0797  24.3         3.44     -6.55  3.73
    Respiration = a0 e a1(T-a2), where T = soil temperature (ºC) at 5 cm below the surface.
    Photosynthesis = b2 – (b0Q)/(Q+b1), where

         Q = incoming photosynthetically active radiation (mol m-2 s-1)

                                     Figure Legends

Figure []. Meteorological observations at or near the Willow Creek flux
tower: a) Maximum daily photosynthetically active radiation (PAR); b) mean daily air
temperature at 29.6 m above the soil surface (solid line), and mean daily soil
temperature at 5 cm below the soil surface (dashed line); c) mean daytime vapor
pressure deficit at the canopy top (20.3 m); d) mean water equivalent precipitation rates;
e) mean daily soil moisture content at 20 cm below the soil surface (solid line), and
mean daily soil water integrated to 1 m below the soil surface; and f) mean daily snow

Figure []. Nonscreened net ecosystem exchange (NEE) from 1999-
2000, showing exceptionally large respiration fluxes associated with winds from 50 to

Figure []. Reduction in net ecosystem exchange (NEE) associated with u*
measurements less than 0.1 m s-1 (1999-2000 data excluding observations when wind
directions were from 50 to 225)

Figure []. Relationship between large respiration fluxes and
temperature inversion at the canopy (difference in air temperature at 29.6 and 18.3 m)
when wind directions are from 50 to 225 and u* is greater than 0.1 m s-1.

Figure []. Nearly complete energy closure for nonscreened 1999-2000 Willow
Creek flux measurements.

Figure []. Estimates of partitioning between photosynthesis and respiration
using gap-filled data from 2000 (mean  standard deviation for each day of the month).

Figure []. Daily accumulation of carbon by the forest stand at Willow
Creek during 2000 (plus sign=observation, triangle=gap-filled using PAR and soil T
relationships, square=gap-filled using seasonal diurnal average).

Figure []. Seasonal changes in daily integrated sensible and latent heat
fluxes at Willow Creek from 1999 to 2000.

Figure []. Seasonal changes in integrated midday (1000 to 1400 CDT)
Bowen ratio and the Priestley-Taylor constant, =ETActual/ETPotential.

Figure []. Lack of transpiration index response due to soil water storage (half-
hour midday measurements for June, July and August, 1999-2000), suggesting that the
trees at Willow Creek are well supplied with water.

Figure [ or]. Lack of transpiration index reponse due to incoming
radiation or vapor pressure deficit at the canopy surface (half-hour midday

measurements for June, July and August, 1999-2000), providing no evidence for
stomatal closure on evapotranspiration


To top