Transcript Slide 1
Prairie Ridge Soil Profile
Amy Keyworth
Jovi Saquing
November 2006
Outline
• What we expect to see
• Why
• What we do see
What we expect to see
•
•
•
•
13C – increase with depth
C/N – decrease with depth
% C – decrease with depth
% N – decrease with depth
What we expect to see
Fig 2 middle. Carbon isotopic composition profiles.
Undisturbed site
Disturbed (agricultural) site
(J.G. Wynn, et al., 2006)
What we expect to see
Fig 2. Carbon concentration profiles.
Undisturbed site
Disturbed (agricultural) site
“Kink” in the Cz curve reflects root depth or productivity zone
(J.G. Wynn, et al., 2006)
Why we expect to see it
•
•
•
•
Suess effect
Soil carbon mixing
Preferential microbial decomposition
Kinetic fractionation
Why we expect to see it
•
Suess effect –
– Older, deeper SOM originated when
atmospheric 13C was more positive (CO2
was heavier)
– From 1744 to 1993, difference in 13C app 1.3 ‰
– Typical soil profile differences = 3 ‰
Suess effect
Fig. 1A. Mixing of SOC derived from the modern atmosphere
versus that derived from a pre-Industrial Revolution
atmosphere. (J.G. Wynn, et al., 2006)
Why we expect to see it
•
Soil carbon mixing
– Surface litter (depleted) vs. root derived
(enriched) SOM
– Variable biomass inputs (C3 vs. C4 plants)
– Some of the carbon incorporated into SOM
by these critters has an atmospheric, not
SOM source.
– Atmospheric C is heavier. Atmospheric CO2
in the soil is 4.4 ‰ heavier than CO2
metabolized by decomposition (Wedin,
1995)
Soil carbon mixing
- Surface litter (depleted) vs. root derived (enriched) SOM
Fig. 1B. Mixing of leaf litter-derived SOC and root-derived
SOC. (J.G. Wynn, et al., 2006)
Soil carbon mixing
- Variable biomass inputs (C3 vs. C4 plants)
Fig. 1C. Mixing of SOC formed under two different vegetation
communities (e.g. C3 vs C4)(slope could vary from positive to
negative depending on direction of shift). (J.G. Wynn, et al., 2006)
Why we expect to see it
•
Preferential microbial decomposition
– Lipids, lignin, cellulose - 13C depleted with
respect to whole plant
– Sugars, amino acids, hemicellulose, pectin 13C enriched
– Lipids and lignin are preferentially
accumulated in early decomposition
– Works against soil depth enrichment
Why we expect to see it
•
Kinetic fractionation
– Microbes choose lighter C
– Microbial respiration of CO2 – 12C
preferentially respired
– Frequently use Rayleigh distillation analyses
(Wynn 2006)
– No direct evidence for this (Ehleringer 2000)
– Preferential preservation of 13C enriched
decomposition products of microbial
transformation
Kinetic fractionation
Fig. 1D. 13C distillation during decomposing SOM. The gray
lines show the model with varying fractionation factors from
0.997 to 0.999. (J.G. Wynn, et al., 2006)
Kinetic fractionation
Rayleigh distillation
13Cf
1
1000
1
F
13
Ci
1
e
t
1
1
1000
e 1t 1 t 1
•
•
•
•
•
•
fraction of remaining soil organic matter (SOC) – approximated by the
calculated value of fSOC
13Cf isotopic composition of SOC when sampled
13Ci isotopic composition of input from biomass
α
fractionation factor between SOC and respired CO2
e
efficiency of microbial assimilation
t
fraction of assimilated carbon retained by a stabilized pool of SOM
F
Assumptions by Wynn etal
•
Open system
– All components decompose
– Contribute to soil-respired CO2 at same rate with depth
•
FSOC fSOC
Anthropogenic mixing (agriculture)
Wynn fig 9 – various reasons that
disturbed land might not conform to
nice regression curve in fig 1D
A – natural
B – introduce C4 plants, enriched in
13C
C – Cropping – removes new, low
13C material, leading to surface
enrichment
D – Erosion – removes upper layer,
moving the whole curve up
E – Reintroduce soil organic carbon
(better management practices) –
reverses the trends in C, D, and E
Order of decay of compounds
(Melillo 1989)
1. Loss of C fractions depleted in 13C (p 192)
–
–
–
–
tannins
non-polar components
water-soluble compounds
lignin, also depleted in 13C is conserved
2. Cellulose – C fraction enriched in 13C
3. Lignin
– Recalcitrant
– Can be enhanced by addition of simple sugars
– N may slow lignin decay (Fenn etal 1981, Keyser etal
1978) – not proven
Controls on decay
Melillo, et al, 1989
•
•
•
•
Temperature
Moisture
Soil texture
Availability of labile C and N
What we do see - results
• 13C – increase 0.4 ‰ with to 8 cm (PRS
18 = anomaly)
• C/N – increases to 8 cm, then decreases
• % C – decrease with depth (PRS 15 =
anomaly)
• % N – decrease with depth (PRS 15 =
anomaly)
What we do see - 13C
Depth vs delta 13C
delta 13C
-25
-20
-15
0
2
Depth (cm)
4
6
8
10
12
14
16
Increase 0.4 ‰ with to 8 cm (PRS 18 = anomaly)
What we do see - C/N
Depth vs C/N
C/N
0
5
10
15
20
0
2
Depth (cm)
4
6
8
10
12
14
16
Increases to 8 cm, then decreases
What we do see - % C
Depth vs %C
%C
0
2
4
0
2
Depth (cm)
4
6
8
10
12
14
16
Decrease with depth (PRS 15 = anomaly)
What we do see - % N
Depth vs %N
%N
0
0.2
0.4
0
2
Depth (cm)
4
6
8
10
12
14
16
Decrease with depth (PRS 15 = anomaly)
Soil FTIR (normalized)
Soil FTIR Normalized
100
90
80
50
40
absorbance
60
Absorbance
70
30
20
10
4000
3500
4500
3000
4000
3500
2500
3000
2000
2500
2000
1500
1500
1000
wave number
number
Wavewave
number
(cm-1)
15
15
16
16
17
18
17
7
18
7
1000
500
0
0
500
FTIR results
• PRS 7 and PRS 15, both surface soils,
have similar absorbencies
• All soils have peak at wavelength 1032
• All 5 spectra have similar peaks, though
not necessarily similar absorbencies
• In our bulk and heavy samples, are the
mineral spectra masking the organics, as
in Poirier’s M-SOM?
0
3000
2500
2000
1500
1000
500
0
Soil FTIR (normalized)
wave number
Soil FTIR Normalized
100
15
16
17
18
7
90
80
50
absorbance
60
40
30
20
10
0
4500
4000
3500
3000
2500
2000
1500
1000
500
0
wave number
Wave
number (cm-1)
15
16
17
18
7
•PRS 7 and PRS 15, both surface soils, have similar absorbencies
•All soils have peak at wavelength 1032
•All 5 spectra have similar peaks, though not necessarily similar absorbencies
•In our bulk and heavy samples, are the mineral spectra masking the organics,
as in Poirier’s M-SOM?
Absorbance
70
•PRS 7 and PRS 15, both surface soils, have similar absorbencies
•All soils have peak at wavelength 1032
•All 5 spectra have similar peaks, though not necessarily similar absorbencies
•In our bulk and heavy samples, are the mineral spectra masking the organics,
as in Poirier’s M-SOM?