Monitoring fluvial pollen transport, its relationship to catchment vegetation and implications for palaeoenvironmental studies

Despite being the most important source of pollen and spore input into most lakes and near-shore marine sediments, we know very little about fluvial (waterborne) pollen and spore transport. This paper presents the results of a dedicated monitoring programme conducted over 2 years and at a catchment scale in South West England. The land use of the nine sub-catchments monitored was determined using Landsat Thematic Data. At two stations, pollen and spore sampling through storm hydrographs was undertaken whilst at the other 7 sub-catchments only peak flow samples were collected. Samples were also collected from re-suspended bed material, riverbanks and at low flows. Airborne pollen flux was monitored using modified Tauber traps. The results support previous research illustrating how the vast majority of fluvial pollen and spores are transported during floods (in this case 9 1 %) and that the main control on waterbome pollen and spore assemblages is the catchment vegetation. However, strong seasonal effects are shown as well as the importance of distinctive sources, such as the riparian input, bed re-suspension and overland flow into drains and tributaries. Fine sediment in river pools appears to act as a selective store of damaged cereal-type pollen grains in arable catchments and this can reduce the inherent underestimate of arable land from pollen diagrams with a high fluvial input and increase the visibility of early agriculture. in order to simulate the likely result of a flood-dominated influx to a small lake scenario, modelling was undertaken whereby different sub-catchments were substituted in order to represent changes in catchment vegetation under a constant hydrological regime. The results show the dampened response of land use groups to catchment land use change, and the frequent occurrence of anomalous single-level peaks due to seasonal flushes from specific near-stream vegetation types. Both these features are commonly seen in lake pollen diagrams. Fluvial pollen and spore loading is dependant upon discharge and so concentrations in laminated or varved sediments could be regarded as a proxy for flood magnitude. The implications for this study on the interpretation of lake and near-shore marine pollen and spore diagrams are discussed and it is argued that a more quantitative approach to waterbome pollen could improve the estimation of land use from lakes in the temperate zone. (C) 2007 Elsevier B.V. All rights reserved.