Wood and soil surface CO2 flux from the Tapajós National ForestEvilene C. Lopes1, Michael Keller1,2, Patrick M. Crill1,3, Ruth K. Varner4, William Z. de Mello5
1CSRC Morse Hall, University of New Hampshire, Durham, NH, 03824, E-mail: [email protected] Forest Service, E-mail: [email protected]; , E-mail: [email protected];
4CCRC UNH, E-mail: [email protected]; 5Universidade Federal Fluminense, E-mail: [email protected]
Diameter at breast height (cm)
0 20 40 60 80 100
CO
2 f
lux
(um
ol m
-2 s
-1)
0
1
2
3
4
5
6
7
8
AB = 4 mb ln (dmax/dmin) / [ (dmax– dmin)]
SUMMARY
The exchange of carbon dioxide (CO2) between tropical forests and the
atmosphere is an important component of the global carbon cycle. A better
understanding of controlling mechanisms and magnitude of CO2 sources from tropical
forests will improve our ability to understand the global carbon budget.
Stem CO2 fluxes were measured using dynamic flux enclosures (Figure 3a
and b). Annual averages at TNF averaged 1.7 mol m-2 s-1. Wood surface area was
measured (4161 m2 ha-1) and wood CO2 flux was extrapolated to ground area resulting
in an annual flux of 259 gC m-2 yr-1, corresponding to about 8% of the gross primary
productivity.
Line sampling of soil-atmosphere CO2 fluxes were made on randomly placed
30 m transects at TNF (Figure 3 a and c). Annually averaged fluxes were 4.7 + 0.2
mol m-2 s-1. Average soil CO2 flux during the wet season was higher (4.9 + 0.3 mol
m-2 s-1) than during the dry season (4.4 + 0.2 mol m-2 s-1). Geostatistical analysis of
grid sampling of soil CO2 fluxes indicated that they were not spatially dependent down
to 15 m separation distance (Figure 3d.). The estimated annual average of soil
surface CO2 flux for the TNF was 1780 gC m-2 yr-1.
ACKNOWLEDGEMENTS
We thank NASA for financial support. LBA office in Santarém, Sr. Antonio Leite (Treviso), for the help working in the selectively logging area at the TNF. Francisco Alves, Cleuton Pereira, Rosenildes Santos, Willey Machado, Hudson Silva, Jadson Dias, Eráclito Neto assistance in the field. Peter Czepiel helped with geostatistical analysis of soil CO2 flux.
Ste
m C
O2
(um
ol m
-2 s
-1)
0
1
2
3
4
5
6
Emergent Dominant Co-dominant Supressed
FIGURE 3 Portable system for CO2 flux measurement for soil and wood. (A) Back-pack IRGA system; (B) plexiglass dynamic chambers for stems; (C) PVC and ABS plastic dynamic soil chamber; and (D) experiment grid for the spatial analysis of soil CO2 flux at the km 67 site of the Tapajós National Forest.
(B)
(C)
(D)
Separation distance (meters)
0 200 400 600 800 1000 1200 1400 1600
Sem
ivar
ianc
e
0
5
10
15
20
25
30
35
Wet seasonDry season
Surface area
(m2 ha-1)
Stems 2391
Branches 1770
Total 4161
FIGURE 8 Monthly averages (+SE) of soil-atmosphere CO2 flux and soil moisture from line sampling in 2001 through 2004 from an undisturbed forest site (km 67) at the TNF. The annual averages of soil CO2 flux and soil moisture were 4.7 mol m-2 s-1 and 44.1%, respectively.
FIGURE 9 Semivariograms of soil CO2 flux for wet and dry periods at the undisturbed forest site of the TNF. We observed no spatial dependency of soil-atmosphere CO2 fluxes for separation distances greater than 15 m.
FIGURE 5 Surface area of branches was calculated based on the pipe-model theory (Shinozaki et al., 1964). In equation 1mb represents the mass of branches of a trees, dmax is the maximum diameter of branches, dmin is the minimum diameter of branches, is the wood density. The sum of branches and stems equals the wood surface area. Surface area of stems and branches were calculated based on a sample of 20 hectares of forest from the km 67 TNF site (TABLE 2).
FIGURE 1 Stem CO2 fluxes from trees grouped into four canopy positions. Trees located at the higher levels of the canopy (emergent and dominant) presented higher (p<0.0001) average stem CO2 fluxes than trees located at lower levels of the canopy (co-dominant and suppressed).
FIGURE 4. Regression analysis between stem surface area and diameter at breast height for 90 harvested trees at the TNF. Stem surface area was calculated based on the approximation of a frustrum of cone (above).
Diameter at breast height (cm)
0 20 40 60 80 100 120 140 160
Ste
m s
urfa
ce a
rea
(m
2 )
0
20
40
60
80
100
120
140
(A)
Date Q10
01/21/03 1.9
01/22/03 1.2
01/23/03 1.5
08/11/03 3.1
08/12/03 3.4
FIGURE 2 Example of continuous measurement of stem CO2 flux and temperature. The Q10 values for five continuous measurement periods are summarized in Table 1. These values indicate that for an increase of 10oC in temperature stem CO2 flux increase between 1.2 and 3.4 times.
01/01/01 07/01/01 01/01/02 07/01/02 01/01/03 07/01/03 01/01/04 07/01/04
Pre
cip
itatio
n (
mm
day
-1)
0
20
40
60
80
100
120
140
160
Soi
l CO
2 f
lux
(um
ol m
-2 s
-1)
0
2
4
6
8
10
12
Soi
l moi
stur
e (%
)
30
35
40
45
50
55
60
Soil CO2 flux
Soil moisture
TABLE 1
TABLE 2
As = -5.84 + 0.7 DBH
r2 = 0.76, P < 0.0001
As = p (R + r) [(h2 + (R – r)2]1/2
FIGURE 6 Biweekly averages of stem CO2 fluxes (+ SE) from undisturbed and selectively logged forest sites at the TNF. Fluxes measured at the undisturbed forest were not significantly different from those measured at the selectively logged site. There was no obvious seasonal variability.
FIGURE 7 Stem CO2 flux versus diameter at breast height (cm) for trees of undisturbed and selectively logged sites. A significant (p<0.0001) and weak (r2 = 0.2) correlation was observed CO2 flux and DBH for both sites combined.
10/1/01 4/1/02 10/1/02 4/1/03 10/1/03 4/1/04 10/1/04
Pre
cipi
tatio
n (m
m d
ay-1
)
0
20
40
60
80
100
120
140
160
Ste
m C
O2 f
lux
(um
ol m
-2 s
-1)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Undisturbed Selectively logged
CO2 flux = 0.7+0.04(DBH)
r2 = 0.2, p < 0.0001
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
12:0
0 P
M
6:00
PM
12:0
0 A
M
6:00
AM
12:0
0 P
M
6:00
PM
12:0
0 A
M
6:00
AM
12:0
0 P
M
6:00
PM
12:0
0 A
M
6:00
AM
12:0
0 P
M
6:00
PM
Ste
m C
O2
flux
(m
ol m
-2 s
-1)
20
22
24
26
28
30
32
34
36
Ste
m t
empe
ratu
re (
o C)
CO2 Flux
wood T
CO2 Flux
Top Related