The importance of edge effects in line transect censuses applied in marshland habitats

Abstract

We estimated relative densities of the 8 most abundant passerine bird species using the Finnish line transect method, and compared them with the results of territory mapping in an extensive Hungarian marshland. Along the transect route on the dikes we found a high positive edge effect for several passerine species. For density estimation in the edge zone, main belt data (0-25m) proved to be useful, but densities for the whole area were overestimated when calculated directly from main belt data for the Bearded Tit (Panurus biarmicus), the Moustached Warbler (Acrocephalus melanopogon), the Sedge Warbler (Acrocephalus schoenobaenus), and the Marsh Warbler (Acrocephalus palustris) . The Finnish line transect method, especially when the linear model is applied, proved to be useful for estimation of densities of these species using observations from the whole survey area . Species showing little or no edge effect (Savi's Warbler Locustella luscinioides, the Great Reed Warbler Acrocephalus arundinaceus, and the Reed Bunting Emberiza schoeniclus) indicated a tendency for underestimation of density. For these species and for the Reed Warbler Acrocephalus scirpaceus the negative exponential model gave better results than the linear model . We also discuss factors affecting line transect censuses in marshland habitats (edge effect, territory size, singing activity pattern vs . detectability) . Of these factors, the edge effect along the observer's route may be more important than is generally the case in woodland habitats . The optimal date for a census is mid-May, but a separate census in late April would also be helpful when a passerine bird community in central European marshland habitat is surveyed.