Salim Ali School of Ecology and Environmental Sciences, Pondicherry University, Pondicherry 605 014, INDIA.
The primary aim of this study was to understand the relationship between modern vegetation and polled assemblage in the surface samples as an essential step to interpret the fossil pollen spectra.
To achieve this, estimation of vegetation composition was carried out in sites of investigation. The quantitative estimation of the vegetation were established by laying transects and quadrats. The similarities within sites and between sites were obtained by using different measures of indices.
The main conclusions drawn from this study are:
The tree density reported is within the range for low elevation evergreen forests of the Western Ghats (Pascal, 1988). The basal area values reported here are within the range of those reported for other sholas in the Nilgiris as well as low elevation evergreen forests (Pascal, 1988). The Shannon’s index values are lower than those of the low elevation evergreen forests. Similarly, the shola forests are shown to be less diverse than montane forests at comparable altitudes in neotropics (Sukumar et al., 1995).
Jaccard’s and Chord index values of dissimilarity within types are lower than those reported for low elevation forests of the Western Ghats (Utkarsh et al., 1998).
Dendograms based on Jaccard’s and Chord index do not show separate clusters for pear bogs and grasslands. Ordination also do not indicate any demarcation between the two types of their constituent species.
To understand the relation between modern vegetation-pollen assemblage correlations are carried out. Use of numerical techniques are also attempted to establish the relationship between modern pollen assemblages from different sites. The conclusions draw from this study are.
The results indicate that Sandynallah surface samples ae less diverse in pollen assemblage. Poaceae and Cyperaceae show very high proportion. The proportion of arboreal is very low indicating the absence of shola patch close to the basin. Low aquatics could be due to non-availability if soil moisture. Bangitappal and Varahapallam surface samples show better representation in shola elements. Cyperaceae is low in both whereas aquatics is high in Bangitappal and low in Varahapallam.
The results from correlation values indicate that all surface pollen samples have high values with grassland types than peat bogs, and the sholas have lowest values. This shows the presence of abundant grassland taxa in the surface samples. The high value with Varahapallam shola compared to other sholas could be due to the disturbance seen around this patch, such as plantations. This would have allowed the increase of grassland elements along with shola components.
In this study reconstruction of the past vegetation by using peat core samples were carried out by palynological investigations. Numerical techniques are used to understand the relationship between samples across the depths corresponding to different periods of time established by chronologies based on 14C dating of peats.
The result depict that predominance of grasses with other herbaceous elements indicate gradual shift towards humid conditions around 16 kys. BP. A decrease in grasses and increase in arboreal pollen towards 12.5 kyr BP and 10kyr BP explain the expansion of forests towards the Holocene. The period between 5 kyr BP to 2kys BP show decrease in arboreal and increase in Cyperaceae and Poaceae aridity during this period.
The establishment of some of core shola species since -2.5 kyr BP could be suggestive of more humid climate in the montane region. Although human disturbance could have occurred around 2 kyr BP, as the pollen record lack evidence of direct anthropogenic indicators. It is an unlikely explanation for the changes registered in the pollen data. This may also coincide with the evidence obtained from Ethiopia (Bonnefille and Mohammed, 1994) which indicates the upward regrowth of the forest indicative of the recent warming around 3 kyr BP.
Pollen profiles are discussed using well established evidence on fluctuation in climate and vegetation that have been obtained based on chronologies in 14C dates of and δ13C values from peat samples. The main conclusions drawn from this study are the coincidence between high δ13C values and high percentage of Poaceae and Cyperaceae is not observed in the pollen and δ13C records. This could be due to the effect of differential residence times of organic carbon prior to its sedimentation explained in east African rift lakes and peat bogs (Hillaire-Marcel et al., 1989). The comparison of pollen and the δ13C signature (based on Sukumar et al., 1993, Rajagopalan, 1996) broadly corresponds with each other which is also indicated in the detrended correspondence analysis.
Palaeoecology has traditionally been associated with the reconstruction of vegetation and climatic changes in the past. As pollen assemblages being the potential indicators of the underlying vegetation patterns, the study of these provide as base line date about the past ecosystem. Reconstruction of these vegetation patterns helps in assessing in the naturalness, fragility of the ecosystems in response to human activity and other biotic factors such as grazing. Quantitative study of modern vegetation, surface sample and fine resolution sampling of peat cores from a number of spatially separated sites would help in better understanding and interpreting the fossil pollen spectra. This would also help us strengthening other proxy record evidences. It would be interesting to understand the magnitude of human impact on vegetation and climate by sampling of peats from number of sites, spanning at least 2 kyr BP. Study of multidisciplinary approach such as pollen investigation δ13C record and phytolith study from these samples would help us in understanding anthropogenic effects on vegetation and climate in this region. These provide basic information to develop essential strategies for restoration and enhancement of the ecosystem. Increased number of quantified vegetation, surface pollen studies from this region would be f great help in understanding underlying processes of pollen dispersal and sedimentation inti peat bogs. Such studies from a region would be very important in developing models that could help in predicting the future global climatic changes.