3.4.
3
.
13
C-NMR spectroscopy
Fig. 4 presents the
13
C-NMR spectra of the HAs extracted from organic matter
-
rich horizons
.
Two HA groups can be distinguished:
- HAs from the A
1
horizons were marked by the abundance of alkyl C (0-30 ppm), O-alkyl C
(55-57 ppm related to CH
3
-O- groups derived from lignin), N-alkyl groups and an intense
carbonyl C region (170 ppm).
- HAs from the A
3
horizons of the Nilg 16 and Kota 0 pedons were dominated by aromatic C
(130 ppm) and carbonyls (170 ppm), whereas alkyl C (0-30 ppm) bands were virtually absent.
The results confirm those obtained from elemental analysis and IR spectroscopy: aromatic
HAs occurred in depth whereas relatively aliphatic ones were present close to the surface.
Figure 3:
FT-IR spectra of humic acids from the Nilg 16 (A) and Kota 0 (B) pedons
(A
1
and A
3
horizons).
Moreover, the NMR spectra of the humic acids from the sombric-like subsurface A horizons
of the Nilg 19 and Kota 0 pedons presented features that were very similar to those obtained
by Golchin et al. (1997) on charred residues from grasses.

4. Discussion
4.1 Vegetation changes recorded by the soils
In the case of Nilgiris, as in many other tropical areas, C4-type plants generally dominate in
grasslands whereas C3-type plants specifically occur in forests. Nevertheless, C3-type herbs
may have also replaced C4-type grasses in grassland, particularly during the deglaciation
period (18,000-10,000 yr BP), when high moisture conditions were supposed to have
favoured C3-type species (Sukumar et al., 1993). In the present study, the interpretation of
13
C values in terms of plant formations does not pose any problem with regard to the C4-
type plant signatures; it is clearly attributed to grassland formations. For the C3-type plant
signatures, when a surface horizon is concerned, the nature of the current vegetation, forest or
plantation is in agreement with this signature.
Taking the above interpretations into account, the data listed in Table 3 indicate that the four
soils have recorded rather different vegetation successions. For instance, in Nilg 16, a clear
grassland signature shows itself in a layer that is both close to the soil surface (20-35 cm) and
characterized by a recent MRT (826 years BP). In contrast, the first 40 cm of the Nilg 19
epipedon have a SOM clearly deriving from a forest vegetation that already existed at this
Figure 4
: Solid state
13
C-NMR spectra of humic acids from the Nilg 16 (A) and Kota 0 (B) pedons
(A
1
and A
3
horizons).
particular site 2,000 years ago. Thus, our results are obviously too limited and also too
scattered to attempt any reconstitution of the recent history of the vegetation dynamic in the

grassland-forest mosaic of the Nilgiris. They are, however, in reasonable agreement with the


You've reached the end of your free preview.
Want to read all 26 pages?
- Spring '14
- DanielKevles
- The Land, Mercedes-Benz CL-Class, Humus, Howard Staunton