POLLEN ANALYSIS AS A REMOTE BIOLOGICAL SENSOR: USING MELISSOPALYNOLOGY AND SURFACE SOIL DATA FOR AN INTEGRATED LANDSCAPE SCALE VEGETATION ASSESSMENT IN SOUTH INDIA

We use pollen assemblages from two sources, surface soil and bee pollen to characterize modern pollen spectra from contrasting landscapes, evaluate their potential as biological proxies complementing each other in reconstructing vegetation comprised of anemophilous and entomophilous plants. The bee pollen assemblages are from honeycombs and corbicular loads from Apis cerana and A. florea. We try to demonstrate how classical palynology functions as an inherently geospatial sensor, offering a powerful integrative approach to landscape-level ecological analysis and generating baseline data on pollen transport mechanisms that remain less explored in the tropics using a case study in southern India in two landscapes - the Coromandel coast and upper montane Nilgiris. The thirty-four samples compiled regionally under two heads (coast, Nilgiris) and under each region two more heads (surface soil, bee pollen) comprised of 151 pollen taxa. Taken together in both landscapes, the bee polen comprised of 63 pollen taxa and the surface soils comprised 136 taxa. Five taxa (Acacia-t, Asteraceae (Echinate), Poaceae, Sapotaceae and Syzygium-t) were found in good proportions under all heads, 17 under three heads (eg: Mimosa pudica, Grewia, Toddalia), 44 under two heads (eg: Coffea, Olea glandulifera). Eighty-four taxa were found exclusively in only one head distributed as follows: surface soils from the coast (43), Nilgiris (26); bee pollen from the coast (11), Nilgiris (4). Data analyses broadly illustrate the potential to further integrate modern pollen data from diverse sources in the same sites and the same season to help quantify pollen-vegetation relationships more holistically and robustly.